Automatically fetches the latest arXiv papers on VLN · VLA · SLAM · 3D · Embodied AI
This project automatically fetches the latest papers from arXiv based on predefined keywords.
- Each section in the README corresponds to a search keyword.
- Only the most recent papers are kept (up to 100 per keyword).
- Click Watch (👀) on the repo to get daily email notifications.
Last update: 2026-04-07
| Title | Date | Abstract | Comment |
|---|---|---|---|
| LatentPilot: Scene-Aware Vision-and-Language Navigation by Dreaming Ahead with Latent Visual Reasoning | 2026-03-31 | ShowExisting vision-and-language navigation (VLN) models primarily reason over past and current visual observations, while largely ignoring the future visual dynamics induced by actions. As a result, they often lack an effective understanding of the causal relationship between actions and how the visual world changes, limiting robust decision-making. Humans, in contrast, can imagine the near future by leveraging action-dynamics causality, which improves both environmental understanding and navigation choices. Inspired by this capability, we propose LatentPilot, a new paradigm that exploits future observations during training as a valuable data source to learn action-conditioned visual dynamics, while requiring no access to future frames at inference. Concretely, we propose a flywheel-style training mechanism that iteratively collects on-policy trajectories and retrains the model to better match the agent's behavior distribution, with an expert takeover triggered when the agent deviates excessively. LatentPilot further learns visual latent tokens without explicit supervision; these latent tokens attend globally in a continuous latent space and are carried across steps, serving as both the current output and the next input, thereby enabling the agent to dream ahead and reason about how actions will affect subsequent observations. Experiments on R2R-CE, RxR-CE, and R2R-PE benchmarks achieve new SOTA results, and real-robot tests across diverse environments demonstrate LatentPilot's superior understanding of environment-action dynamics in scene. Project page:https://abdd.top/latentpilot/ |
Proje...Project page:https://abdd.top/latentpilot/ |
| Dream to Recall: Imagination-Guided Experience Retrieval for Memory-Persistent Vision-and-Language Navigation | 2026-03-30 | ShowVision-and-Language Navigation (VLN) requires agents to follow natural language instructions through environments, with memory-persistent variants demanding progressive improvement through accumulated experience. Existing approaches for memory-persistent VLN face critical limitations: they lack effective memory access mechanisms, instead relying on entire memory incorporation or fixed-horizon lookup, and predominantly store only environmental observations while neglecting navigation behavioral patterns that encode valuable decision-making strategies. We present Memoir, which employs imagination as a retrieval mechanism grounded by explicit memory: a world model imagines future navigation states as queries to selectively retrieve relevant environmental observations and behavioral histories. The approach comprises: 1) a language-conditioned world model that imagines future states serving dual purposes: encoding experiences for storage and generating retrieval queries; 2) Hybrid Viewpoint-Level Memory that anchors both observations and behavioral patterns to viewpoints, enabling hybrid retrieval; and 3) an experience-augmented navigation model that integrates retrieved knowledge through specialized encoders. Extensive evaluation across diverse memory-persistent VLN benchmarks with 10 distinct testing scenarios demonstrates Memoir's effectiveness: significant improvements across all scenarios, with 5.4% SPL gains on IR2R over the best memory-persistent baseline, accompanied by 8.3x training speedup and 74% inference memory reduction. The results validate that predictive retrieval of both environmental and behavioral memories enables more effective navigation, with analysis indicating substantial headroom (73.3% vs 93.4% upper bound) for this imagination-guided paradigm. |
Accep...Accepted by IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI) |
| Beyond Textual Knowledge-Leveraging Multimodal Knowledge Bases for Enhancing Vision-and-Language Navigation | 2026-03-27 | ShowVision-and-Language Navigation (VLN) requires an agent to navigate through complex unseen environments based on natural language instructions. However, existing methods often struggle to effectively capture key semantic cues and accurately align them with visual observations. To address this limitation, we propose Beyond Textual Knowledge (BTK), a VLN framework that synergistically integrates environment-specific textual knowledge with generative image knowledge bases. BTK employs Qwen3-4B to extract goal-related phrases and utilizes Flux-Schnell to construct two large-scale image knowledge bases: R2R-GP and REVERIE-GP. Additionally, we leverage BLIP-2 to construct a large-scale textual knowledge base derived from panoramic views, providing environment-specific semantic cues. These multimodal knowledge bases are effectively integrated via the Goal-Aware Augmentor and Knowledge Augmentor, significantly enhancing semantic grounding and cross-modal alignment. Extensive experiments on the R2R dataset with 7,189 trajectories and the REVERIE dataset with 21,702 instructions demonstrate that BTK significantly outperforms existing baselines. On the test unseen splits of R2R and REVERIE, SR increased by 5% and 2.07% respectively, and SPL increased by 4% and 3.69% respectively. The source code is available at https://github.com/yds3/IPM-BTK/. |
Main ...Main paper (37 pages). Accepted for publication by the Information Processing and Management,Volume 63,Issue 6,September 2026,104766 |
| SpatialAnt: Autonomous Zero-Shot Robot Navigation via Active Scene Reconstruction and Visual Anticipation | 2026-03-27 | ShowVision-and-Language Navigation (VLN) has recently benefited from Multimodal Large Language Models (MLLMs), enabling zero-shot navigation. While recent exploration-based zero-shot methods have shown promising results by leveraging global scene priors, they rely on high-quality human-crafted scene reconstructions, which are impractical for real-world robot deployment. When encountering an unseen environment, a robot should build its own priors through pre-exploration. However, these self-built reconstructions are inevitably incomplete and noisy, which severely degrade methods that depend on high-quality scene reconstructions. To address these issues, we propose SpatialAnt, a zero-shot navigation framework designed to bridge the gap between imperfect self-reconstructions and robust execution. SpatialAnt introduces a physical grounding strategy to recover the absolute metric scale for monocular-based reconstructions. Furthermore, rather than treating the noisy self-reconstructed scenes as absolute spatial references, we propose a novel visual anticipation mechanism. This mechanism leverages the noisy point clouds to render future observations, enabling the agent to perform counterfactual reasoning and prune paths that contradict human instructions. Extensive experiments in both simulated and real-world environments demonstrate that SpatialAnt significantly outperforms existing zero-shot methods. We achieve a 66% Success Rate (SR) on R2R-CE and 50.8% SR on RxR-CE benchmarks. Physical deployment on a Hello Robot further confirms the efficiency and efficacy of our framework, achieving a 52% SR in challenging real-world settings. |
10 pa...10 pages, 4 figures, 5 tables. Homepage: https://imnearth.github.io/Spatial-X/ |
| DecoVLN: Decoupling Observation, Reasoning, and Correction for Vision-and-Language Navigation | 2026-03-26 | ShowVision-and-Language Navigation (VLN) requires agents to follow long-horizon instructions and navigate complex 3D environments. However, existing approaches face two major challenges: constructing an effective long-term memory bank and overcoming the compounding errors problem. To address these issues, we propose DecoVLN, an effective framework designed for robust streaming perception and closed-loop control in long-horizon navigation. First, we formulate long-term memory construction as an optimization problem and introduce adaptive refinement mechanism that selects frames from a historical candidate pool by iteratively optimizing a unified scoring function. This function jointly balances three key criteria: semantic relevance to the instruction, visual diversity from the selected memory, and temporal coverage of the historical trajectory. Second, to alleviate compounding errors, we introduce a state-action pair-level corrective finetuning strategy. By leveraging geodesic distance between states to precisely quantify deviation from the expert trajectory, the agent collects high-quality state-action pairs in the trusted region while filtering out the polluted data with low relevance. This improves both the efficiency and stability of error correction. Extensive experiments demonstrate the effectiveness of DecoVLN, and we have deployed it in real-world environments. |
16 pa...16 pages, 8 figures, CVPR2026 |
| T-araVLN: Translator for Agricultural Robotic Agents on Vision-and-Language Navigation | 2026-03-26 | ShowAgricultural robotic agents have been becoming useful helpers in a wide range of agricultural tasks. However, they still heavily rely on manual operations or fixed railways for movement. To address this limitation, the AgriVLN method and the A2A benchmark pioneeringly extend Vision-and-Language Navigation (VLN) to the agricultural domain, enabling agents to navigate to the target positions following the natural language instructions. We observe that AgriVLN can effectively understands the simple instructions, but often misunderstands the complex ones. To bridge this gap, we propose the T-araVLN method, in which we build the instruction translator module to translate noisy and mistaken instructions into refined and precise representations. When evaluated on A2A, our T-araVLN successfully improves Success Rate (SR) from 0.47 to 0.63 and reduces Navigation Error (NE) from 2.91m to 2.28m, demonstrating the state-of-the-art performance in the agricultural VLN domain. Code: https://github.com/AlexTraveling/T-araVLN. |
|
| SpatialFly: Geometry-Guided Representation Alignment for UAV Vision-and-Language Navigation in Urban Environments | 2026-03-22 | ShowUAVs play an important role in applications such as autonomous exploration, disaster response, and infrastructure inspection. However, UAV VLN in complex 3D environments remains challenging. A key difficulty is the structural representation mismatch between 2D visual perception and the 3D trajectory decision space, which limits spatial reasoning. To this end, we propose SpatialFly, a geometry-guided spatial representation framework for UAV VLN. Operating on RGB observations without explicit 3D reconstruction, SpatialFly introduces a geometry-guided 2D representation alignment mechanism. Specifically, the geometric prior injection module injects global structural cues into 2D semantic tokens to provide scene-level geometric guidance. The geometry-aware reparameterization module then aligns 2D semantic tokens with 3D geometric tokens through cross-modal attention, followed by gated residual fusion to preserve semantic discrimination. Experimental results show that SpatialFly consistently outperforms state-of-the-art UAV VLN baselines across both seen and unseen environments, reducing NE by 4.03m and improving SR by 1.27% over the strongest baseline on the unseen Full split. Additional trajectory-level analysis shows that SpatialFly produces trajectories with better path alignment and smoother, more stable motion. |
|
| HaltNav: Reactive Visual Halting over Lightweight Topological Priors for Robust Vision-Language Navigation | 2026-03-19 | ShowVision-and-Language Navigation (VLN) is shifting from rigid, step-by-step instruction following toward open-vocabulary, goal-oriented autonomy. Achieving this transition without exhaustive routing prompts requires agents to leverage structural priors. While prior work often assumes computationally heavy 2D/3D metric maps, we instead exploit a lightweight, text-based osmAG (OpenStreetMap Area Graph), a floorplan-level topological representation that is easy to obtain and maintain. However, global planning over a prior map alone is brittle in real-world deployments, where local connectivity can change (e.g., closed doors or crowded passages), leading to execution-time failures. To address this gap, we propose a hierarchical navigation framework HaltNav that couples the robust global planning of osmAG with the local exploration and instruction-grounding capability of VLN. Our approach features an MLLM-based brain module, which is capable of high-level task grounding and obstruction awareness. Conditioned on osmAG, the brain converts the global route into a sequence of localized execution snippets, providing the VLN executor with prior-grounded, goal-centric sub-instructions. Meanwhile, it detects local anomalies via a mechanism we term Reactive Visual Halting (RVH), which interrupts the local control loop, updates osmAG by invalidating the corresponding topology, and triggers replanning to orchestrate a viable detour. To train this halting capability efficiently, we introduce a data synthesis pipeline that leverages generative models to inject realistic obstacles into otherwise navigable scenes, substantially enriching hard negative samples. Extensive experiments demonstrate that our hierarchical framework outperforms several baseline methods without tedious language instructions, and significantly improves robustness for long-horizon vision-language navigation under environmental changes. |
|
| Learning Goal-Oriented Vision-and-Language Navigation with Self-Improving Demonstrations at Scale | 2026-03-18 | ShowGoal-oriented vision-language navigation requires robust exploration capabilities for agents to navigate to specified goals in unknown environments without step-by-step instructions. Existing methods tend to exclusively utilize shortest-path trajectories, lacking effective exploration priors for training navigation agents. To address the above challenges, we present SID, a goal-oriented vision-and-language navigation learning approach with Self-Improving Demonstrations. Specifically, SID learns an initial agent on the shortest-path data sampled from environments and then leverages this agent to generate novel exploration trajectories. The novel rollouts provide demonstrations with stronger exploration strategies to train a better agent, which in turn produces higher-quality agent demonstrations for the next round of training. We show that this iterative self-improving pipeline readily scales to new environments, and the resulting demonstrations are highly transferable, elevating the performance ceiling across a variety of vision-and-language navigation tasks. Extensive experiments demonstrate that SID significantly boosts the exploration capabilities and generalization of navigation agents. The resulting agent achieves new state-of-the-art performance on goal-oriented vision-and-language navigation benchmarks, including REVERIE, SOON as well as strong transferability to object-goal navigation and VLN-CE. It notably achieves a 50.9% success rate on the unseen validation splits of SOON, surpassing prior leading approaches by a margin of 13.9%. |
|
| AgentVLN: Towards Agentic Vision-and-Language Navigation | 2026-03-18 | ShowVision-and-Language Navigation (VLN) requires an embodied agent to ground complex natural-language instructions into long-horizon navigation in unseen environments. While Vision-Language Models (VLMs) offer strong 2D semantic understanding, current VLN systems remain constrained by limited spatial perception, 2D-3D representation mismatch, and monocular scale ambiguity. In this paper, we propose AgentVLN, a novel and efficient embodied navigation framework that can be deployed on edge computing platforms. We formulate VLN as a Partially Observable Semi-Markov Decision Process (POSMDP) and introduce a VLM-as-Brain paradigm that decouples high-level semantic reasoning from perception and planning via a plug-and-play skill library. To resolve multi-level representation inconsistency, we design a cross-space representation mapping that projects perception-layer 3D topological waypoints into the image plane, yielding pixel-aligned visual prompts for the VLM. Building on this bridge, we integrate a context-aware self-correction and active exploration strategy to recover from occlusions and suppress error accumulation over long trajectories. To further address the spatial ambiguity of instructions in unstructured environments, we propose a Query-Driven Perceptual Chain-of-Thought (QD-PCoT) scheme, enabling the agent with the metacognitive ability to actively seek geometric depth information. Finally, we construct AgentVLN-Instruct, a large-scale instruction-tuning dataset with dynamic stage routing conditioned on target visibility. Extensive experiments show that AgentVLN consistently outperforms prior state-of-the-art methods (SOTA) on long-horizon VLN benchmarks, offering a practical paradigm for lightweight deployment of next-generation embodied navigation models. Code: https://github.com/Allenxinn/AgentVLN. |
19pages, 4 figures |
| P$^{3}$Nav: End-to-End Perception, Prediction and Planning for Vision-and-Language Navigation | 2026-03-18 | ShowIn Vision-and-Language Navigation (VLN), an agent is required to plan a path to the target specified by the language instruction, using its visual observations. Consequently, prevailing VLN methods primarily focus on building powerful planners through visual-textual alignment. However, these approaches often bypass the imperative of comprehensive scene understanding prior to planning, leaving the agent with insufficient perception or prediction capabilities. Thus, we propose P$^{3}$Nav, a novel end-to-end framework integrating perception, prediction, and planning in a unified pipeline to strengthen the VLN agent's scene understanding and boost navigation success. Specifically, P$^{3}$Nav augments perception by extracting complementary cues from object-level and map-level perspectives. Subsequently, our P$^{3}$Nav predicts waypoints to model the agent's potential future states, endowing the agent with intrinsic awareness of candidate positions during navigation. Conditioned on these future waypoints, P$^{3}$Nav further forecasts semantic map cues, enabling proactive planning and reducing the strict reliance on purely historical context. Integrating these perceptual and predictive cues, a holistic planning module finally carries out the VLN tasks. Extensive experiments demonstrate that our P$^{3}$Nav achieves new state-of-the-art performance on the REVERIE, R2R-CE, and RxR-CE benchmarks. |
|
| Trajectory-Diversity-Driven Robust Vision-and-Language Navigation | 2026-03-16 | ShowVision-and-Language Navigation (VLN) requires agents to navigate photo-realistic environments following natural language instructions. Current methods predominantly rely on imitation learning, which suffers from limited generalization and poor robustness to execution perturbations. We present NavGRPO, a reinforcement learning framework that learns goal-directed navigation policies through Group Relative Policy Optimization. By exploring diverse trajectories and optimizing via within-group performance comparisons, our method enables agents to distinguish effective strategies beyond expert paths without requiring additional value networks. Built on ScaleVLN, NavGRPO achieves superior robustness on R2R and REVERIE benchmarks with +3.0% and +1.71% SPL improvements in unseen environments. Under extreme early-stage perturbations, we demonstrate +14.89% SPL gain over the baseline, confirming that goal-directed RL training builds substantially more robust navigation policies. Code and models will be released. |
17pages, 5 figures |
| EmergeNav: Structured Embodied Inference for Zero-Shot Vision-and-Language Navigation in Continuous Environments | 2026-03-16 | ShowZero-shot vision-and-language navigation in continuous environments (VLN-CE) remains challenging for modern vision-language models (VLMs). Although these models encode useful semantic priors, their open-ended reasoning does not directly translate into stable long-horizon embodied execution. We argue that the key bottleneck is not missing knowledge alone, but missing an execution structure for organizing instruction following, perceptual grounding, temporal progress, and stage verification. We propose EmergeNav, a zero-shot framework that formulates continuous VLN as structured embodied inference. EmergeNav combines a Plan--Solve--Transition hierarchy for stage-structured execution, GIPE for goal-conditioned perceptual extraction, contrastive dual-memory reasoning for progress grounding, and role-separated Dual-FOV sensing for time-aligned local control and boundary verification. On VLN-CE, EmergeNav achieves strong zero-shot performance using only open-source VLM backbones and no task-specific training, explicit maps, graph search, or waypoint predictors, reaching 30.00 SR with Qwen3-VL-8B and 37.00 SR with Qwen3-VL-32B. These results suggest that explicit execution structure is a key ingredient for turning VLM priors into stable embodied navigation behavior. |
|
| \textsc{NaVIDA}: Vision-Language Navigation with Inverse Dynamics Augmentation | 2026-03-16 | ShowVision-and-Language Navigation (VLN) requires agents to interpret natural language instructions and act coherently in visually rich environments. However, most existing methods rely on reactive state-action mappings without explicitly action-grounded visual dynamics modeling. Lacking awareness of how actions transform subsequent visual observations, agents cannot plan actions rationally, leading to unstable behaviors, weak generalization, and cumulative error along trajectory. To address these issues, we introduce \textsc{NaVIDA} (\textbf{Nav}igation with \textbf{I}nverse \textbf{D}ynamics \textbf{A}ugmentation), a lightweight VLN framework that incorporates inverse dynamics supervision (IDS) as an explicit objective to embed action-grounded visual dynamics into policy learning. By jointly optimizing this visual dynamics with instruction-conditioned action prediction in a shared representation and action space, \textsc{NaVIDA} provides additional structured supervision that regularizes learning and leads to more stable and consistent navigation. To structure this supervision and extend the effective planning range, \textsc{NaVIDA} employs hierarchical probabilistic action chunking (HPAC), which organizes trajectories into multi-step chunks and provides discriminative, longer-range visual-change cues. Extensive experiments show that \textsc{NaVIDA} achieves superior navigation performance compared to state-of-the-art methods with fewer parameters (3B vs. 8B). Real-world robot evaluations further validate the practical feasibility and effectiveness of our approach. |
27 pages, 11 figures |
| HiMemVLN: Enhancing Reliability of Open-Source Zero-Shot Vision-and-Language Navigation with Hierarchical Memory System | 2026-03-16 | ShowLLM-based agents have demonstrated impressive zero-shot performance in vision-language navigation (VLN) tasks. However, most zero-shot methods primarily rely on closed-source LLMs as navigators, which face challenges related to high token costs and potential data leakage risks. Recent efforts have attempted to address this by using open-source LLMs combined with a spatiotemporal CoT framework, but they still fall far short compared to closed-source models. In this work, we identify a critical issue, Navigation Amnesia, through a detailed analysis of the navigation process. This issue leads to navigation failures and amplifies the gap between open-source and closed-source methods. To address this, we propose HiMemVLN, which incorporates a Hierarchical Memory System into a multimodal large model to enhance visual perception recall and long-term localization, mitigating the amnesia issue and improving the agent's navigation performance. Extensive experiments in both simulated and real-world environments demonstrate that HiMemVLN achieves nearly twice the performance of the open-source state-of-the-art method. The code is available at https://github.com/lvkailin0118/HiMemVLN. |
9 pages, 7 figures |
| All-day Multi-scenes Lifelong Vision-and-Language Navigation with Tucker Adaptation | 2026-03-15 | ShowDeploying vision-and-language navigation (VLN) agents requires adaptation across diverse scenes and environments, but fine-tuning on a specific scenario often causes catastrophic forgetting in others, which severely limits flexible long-term deployment. We formalize this challenge as the all-day multi-scenes lifelong VLN (AML-VLN) problem. Existing parameter-efficient adapters (e.g., LoRA and its variants) are limited by their two-dimensional matrix form, which fails to capture the multi-hierarchical navigation knowledge spanning multiple scenes and environments. To address this, we propose Tucker Adaptation (TuKA), which represents the multi-hierarchical navigation knowledge as a high-order tensor and leverages Tucker decomposition to decouple the knowledge into shared subspaces and scenario-specific experts. We further introduce a decoupled knowledge incremental learning strategy to consolidate shared subspaces while constraining specific experts for decoupled lifelong learning. Building on TuKA, we also develop a VLN agent named AlldayWalker, which continually learns across multiple navigation scenarios, achieving all-day multi-scenes navigation. Extensive experiments show that AlldayWalker consistently outperforms state-of-the-art baselines. |
ICLR 2026 |
| ProFocus: Proactive Perception and Focused Reasoning in Vision-and-Language Navigation | 2026-03-15 | ShowVision-and-Language Navigation (VLN) requires agents to accurately perceive complex visual environments and reason over navigation instructions and histories. However, existing methods passively process redundant visual inputs and treat all historical contexts indiscriminately, resulting in inefficient perception and unfocused reasoning. To address these challenges, we propose \textbf{ProFocus}, a training-free progressive framework that unifies \underline{Pro}active Perception and \underline{Focus}ed Reasoning through collaboration between large language models (LLMs) and vision-language models (VLMs). For proactive perception, ProFocus transforms panoramic observations into structured ego-centric semantic maps, enabling the orchestration agent to identify missing visual information needed for reliable decision-making, and to generate targeted visual queries with corresponding focus regions that guide the perception agent to acquire the required observations. For focused reasoning, we propose Branch-Diverse Monte Carlo Tree Search (BD-MCTS) to identify top-$k$ high-value waypoints from extensive historical candidates. The decision agent focuses reasoning on the historical contexts associated with these waypoints, rather than considering all historical waypoints equally. Extensive experiments validate the effectiveness of ProFocus, achieving state-of-the-art performance among zero-shot methods on R2R and REVERIE benchmarks. |
Accep...Accepted by CVPR 2026 |
| FreeFly-Thinking : Aligning Chain-of-Thought Reasoning with Continuous UAV Navigation | 2026-03-11 | ShowVision-Language Navigation aims to enable agents to understand natural language instructions and carry out appropriate navigation actions in real-world environments. Most work focuses on indoor settings, with little research in complex outdoor scenes. Current UAV Vision-and-Language Navigation models typically act as black boxes without explicit reasoning. We introduce FreeFly-thinking, an end-to-end VLN framework that converts the UAV agent's egocentric images and language instructions into a series of actions, inspired by environment of urban architecture proposed by OpenFly. We first construct a UAV dataset for navigation task, and then performing natural language chain of thought. We adopt a two-stage training strategy: Supervised fine-tuning and Reinforcement fine-tuning. Experiments on unseen test demonstrate a strong performance, presenting robustness and efficiency in UAV navigation issue. |
10 pa...10 pages, 5 figures, ECCV review |
| VLN-Cache: Enabling Token Caching for VLN Models with Visual/Semantic Dynamics Awareness | 2026-03-10 | ShowVision-and-Language Navigation (VLN) increasingly relies on large vision-language models, but their inference cost conflicts with real-time deployment. Token caching is a promising training-free strategy that avoids redundant computation by reusing stable visual tokens across frames. However, existing methods assume a static camera and fixed semantic focus, assumptions that VLN fundamentally violates. We identify two failure modes: (1) visual dynamics, where viewpoint shift displaces token positions across frames, causing position-wise matching to pair misaligned content; (2) semantic dynamics, where token relevance shifts across task stages as navigation progresses, making cached states stale. We propose VLN-Cache, a visual-dynamic-aware and semantic-dynamic-aware caching framework that introduces view-aligned remapping to recover geometric correspondences and a task-relevance saliency filter to veto reuse at semantic transitions. A layer-adaptive entropy policy further balances the per-layer reuse budget. Experiments on the R2R-CE simulation benchmark show up to 1.52x speedup while maintaining competitive navigation success rates. |
|
| Implicit Geometry Representations for Vision-and-Language Navigation from Web Videos | 2026-03-10 | ShowVision-and-Language Navigation (VLN) has long been constrained by the limited diversity and scalability of simulator-curated datasets, which fail to capture the complexity of real-world environments. To overcome this limitation, we introduce a large-scale video-instruction framework derived from web-based room tour videos, enabling agents to learn from natural human walking demonstrations in diverse, realistic indoor settings. Unlike existing datasets, our framework integrates both open-ended description-enriched trajectories and action-enriched trajectories reconstructed in 3D, providing richer spatial and semantic supervision. A key extension in this work is the incorporation of implicit geometry representations, which extract spatial cues directly from RGB frames without requiring fragile 3D reconstruction. This approach substantially improves data utilization, alleviates reconstruction failures, and unlocks large portions of previously unusable video data. Comprehensive experiments across multiple VLN benchmarks (CVDN, SOON, R2R, and REVERIE) demonstrate that our method not only sets new state-of-the-art performance but also enables the development of robust zero-shot navigation agents. By bridging large-scale web videos with implicit spatial reasoning, this work advances embodied navigation towards more scalable, generalizable, and real-world applicable solutions. |
Exten...Extension of CVPR 2025 RoomTour3D with implicit geometric representations |
| CMMR-VLN: Vision-and-Language Navigation via Continual Multimodal Memory Retrieval | 2026-03-09 | ShowAlthough large language models (LLMs) are introduced into vision-and-language navigation (VLN) to improve instruction comprehension and generalization, existing LLM- based VLN lacks the ability to selectively recall and use relevant priori experiences to help navigation tasks, limiting their performance in long-horizon and unfamiliar scenarios. In this work, we propose CMMR-VLN (Continual Multimodal Memory Retrieval based VLN), a VLN framework that endows LLM agents with structured memory and reflection capabilities. Specifically, the CMMR-VLN constructs a multimodal experi- ence memory indexed by panoramic visual images and salient landmarks to retrieve relevant experiences during navigation, introduces a retrieved-augmented generation pipeline to mimick how experienced human navigators leverage priori knowledge, and incorporates a reflection-based memory update strategy that selectively stores complete successful paths and the key initial mistake in failure cases. Comprehensive tests illustrate average success rate improvements of 52.9%, 20.9% and 20.9%, and 200%, 50% and 50% over the NavGPT, the MapGPT, and the DiscussNav in simulation and real tests, respectively eluci- dating the great potential of the CMMR-VLN as a backbone VLN framework. |
|
| LaViRA: Language-Vision-Robot Actions Translation for Zero-Shot Vision Language Navigation in Continuous Environments | 2026-03-04 | ShowLaViRA: Zero-shot Vision-and-Language Navigation in Continuous Environments (VLN-CE) requires an agent to navigate unseen environments based on natural language instructions without any prior training. Current methods face a critical trade-off: either rely on environment-specific waypoint predictors that limit scene generalization, or underutilize the reasoning capabilities of large models during navigation. We introduce LaViRA, a simple yet effective zero-shot framework that addresses this dilemma by decomposing action into a coarse-to-fine hierarchy: Language Action for high-level planning, Vision Action for middle-level perceptual grounding, and Robot Action for low-level control. This modular decomposition allows us to leverage the distinct strengths of different scales of Multimodal Large Language Models (MLLMs) at each stage, creating a system that is powerful in its reasoning, grounding and practical control. LaViRA significantly outperforms existing state-of-the-art methods on the VLN-CE benchmark, demonstrating superior generalization capabilities in unseen environments, while maintaining transparency and efficiency for real-world deployment. Project page: https://robo-lavira.github.io/lavira-zs-vln/ |
ICRA 2026 |
| Endowing Embodied Agents with Spatial Reasoning Capabilities for Vision-and-Language Navigation | 2026-03-02 | ShowEnhancing the spatial perception capabilities of mobile robots is crucial for achieving embodied Vision-and-Language Navigation (VLN). Although significant progress has been made in simulated environments, directly transferring these capabilities to real-world scenarios often results in severe hallucination phenomena, causing robots to lose effective spatial awareness. To address this issue, we propose BrainNav, a bio-inspired spatial cognitive navigation framework inspired by biological spatial cognition theories and cognitive map theory. BrainNav integrates dual-map (coordinate map and topological map) and dual-orientation (relative orientation and absolute orientation) strategies, enabling real-time navigation through dynamic scene capture and path planning. Its five core modules-Hippocampal Memory Hub, Visual Cortex Perception Engine, Parietal Spatial Constructor, Prefrontal Decision Center, and Cerebellar Motion Execution Unit-mimic biological cognitive functions to reduce spatial hallucinations and enhance adaptability. Validated in a zero-shot real-world lab environment using the Limo Pro robot, BrainNav, compatible with GPT-4, outperforms existing State-of-the-Art (SOTA) Vision-and-Language Navigation in Continuous Environments (VLN-CE) methods without fine-tuning. |
|
| Aerial Vision-Language Navigation with a Unified Framework for Spatial, Temporal and Embodied Reasoning | 2026-02-25 | ShowAerial Vision-and-Language Navigation (VLN) aims to enable unmanned aerial vehicles (UAVs) to interpret natural language instructions and navigate complex urban environments using onboard visual observation. This task holds promise for real-world applications such as low-altitude inspection, search-and-rescue, and autonomous aerial delivery. Existing methods often rely on panoramic images, depth inputs, or odometry to support spatial reasoning and action planning. These requirements increase system cost and integration complexity, thus hindering practical deployment for lightweight UAVs. We present a unified aerial VLN framework that operates solely on egocentric monocular RGB observations and natural language instructions. The model formulates navigation as a next-token prediction problem, jointly optimizing spatial perception, trajectory reasoning, and action prediction through prompt-guided multi-task learning. Moreover, we propose a keyframe selection strategy to reduce visual redundancy by retaining semantically informative frames, along with an action merging and label reweighting mechanism that mitigates long-tailed supervision imbalance and facilitates stable multi-task co-training. Extensive experiments on the AerialVLN and OpenFly benchmark validate the effectiveness of our method. Under the challenging monocular RGB-only setting, our model achieves strong results across both seen and unseen environments. It significantly outperforms existing RGB-only baselines and narrows the performance gap with state-of-the-art panoramic RGB-D counterparts. Comprehensive ablation studies further demonstrate the contribution of our task design and architectural choices. |
Under...Under Review, 15 pages, 11 figures |
| JanusVLN: Decoupling Semantics and Spatiality with Dual Implicit Memory for Vision-Language Navigation | 2026-02-25 | ShowVision-and-Language Navigation requires an embodied agent to navigate through unseen environments, guided by natural language instructions and a continuous video stream. Recent advances in VLN have been driven by the powerful semantic understanding of Multimodal Large Language Models. However, these methods typically rely on explicit semantic memory, such as building textual cognitive maps or storing historical visual frames. This type of method suffers from spatial information loss, computational redundancy, and memory bloat, which impede efficient navigation. Inspired by the implicit scene representation in human navigation, analogous to the left brain's semantic understanding and the right brain's spatial cognition, we propose JanusVLN, a novel VLN framework featuring a dual implicit neural memory that models spatial-geometric and visual-semantic memory as separate, compact, and fixed-size neural representations. This framework first extends the MLLM to incorporate 3D prior knowledge from the spatial-geometric encoder, thereby enhancing the spatial reasoning capabilities of models based solely on RGB input. Then, the historical key-value caches from the spatial-geometric and visual-semantic encoders are constructed into a dual implicit memory. By retaining only the KVs of tokens in the initial and sliding window, redundant computation is avoided, enabling efficient incremental updates. Extensive experiments demonstrate that JanusVLN outperforms over 20 recent methods to achieve SOTA performance. For example, the success rate improves by 10.5-35.5 compared to methods using multiple data types as input and by 3.6-10.8 compared to methods using more RGB training data. This indicates that the proposed dual implicit neural memory, as a novel paradigm, explores promising new directions for future VLN research. Ours project page: https://miv-xjtu.github.io/JanusVLN.github.io/. |
Accep...Accepted to ICLR 2026. Project page: https://miv-xjtu.github.io/JanusVLN.github.io/ |
| Learning to Retrieve Navigable Candidates for Efficient Vision-and-Language Navigation | 2026-02-17 | ShowVision-and-Language Navigation (VLN) requires an agent to follow natural-language instructions and navigate through previously unseen environments. Recent approaches increasingly employ large language models (LLMs) as high-level navigators due to their flexibility and reasoning capability. However, prompt-based LLM navigation often suffers from inefficient decision-making, as the model must repeatedly interpret instructions from scratch and reason over noisy and verbose navigable candidates at each step. In this paper, we propose a retrieval-augmented framework to improve the efficiency and stability of LLM-based VLN without modifying or fine-tuning the underlying language model. Our approach introduces retrieval at two complementary levels. At the episode level, an instruction-level embedding retriever selects semantically similar successful navigation trajectories as in-context exemplars, providing task-specific priors for instruction grounding. At the step level, an imitation-learned candidate retriever prunes irrelevant navigable directions before LLM inference, reducing action ambiguity and prompt complexity. Both retrieval modules are lightweight, modular, and trained independently of the LLM. We evaluate our method on the Room-to-Room (R2R) benchmark. Experimental results demonstrate consistent improvements in Success Rate, Oracle Success Rate, and SPL on both seen and unseen environments. Ablation studies further show that instruction-level exemplar retrieval and candidate pruning contribute complementary benefits to global guidance and step-wise decision efficiency. These results indicate that retrieval-augmented decision support is an effective and scalable strategy for enhancing LLM-based vision-and-language navigation. |
|
| One Agent to Guide Them All: Empowering MLLMs for Vision-and-Language Navigation via Explicit World Representation | 2026-02-17 | ShowA navigable agent needs to understand both high-level semantic instructions and precise spatial perceptions. Building navigation agents centered on Multimodal Large Language Models (MLLMs) demonstrates a promising solution due to their powerful generalization ability. However, the current tightly coupled design dramatically limits system performance. In this work, we propose a decoupled design that separates low-level spatial state estimation from high-level semantic planning. Unlike previous methods that rely on predefined, oversimplified textual maps, we introduce an interactive metric world representation that maintains rich and consistent information, allowing MLLMs to interact with and reason on it for decision-making. Furthermore, counterfactual reasoning is introduced to further elicit MLLMs' capacity, while the metric world representation ensures the physical validity of the produced actions. We conduct comprehensive experiments in both simulated and real-world environments. Our method establishes a new zero-shot state-of-the-art, achieving 48.8% Success Rate (SR) in R2R-CE and 42.2% in RxR-CE benchmarks. Furthermore, to validate the versatility of our metric representation, we demonstrate zero-shot sim-to-real transfer across diverse embodiments, including a wheeled TurtleBot 4 and a custom-built aerial drone. These real-world deployments verify that our decoupled framework serves as a robust, domain-invariant interface for embodied Vision-and-Language navigation. |
|
| UNeMo: Collaborative Visual-Language Reasoning and Navigation via a Multimodal World Model | 2026-02-07 | ShowVision-and-Language Navigation (VLN) requires agents to autonomously navigate complex environments via visual images and natural language instructions--remains highly challenging. Recent research on enhancing language-guided navigation reasoning using pre-trained large language models (LLMs) has shown promising prospects. However, the reasoning of such methods is limited to the linguistic modality, lacking visual reasoning capabilities. Moreover, existing reasoning modules are optimized separately from navigation policies, leading to incompatibility and potential conflicts in optimization objectives.To tackle these challenges, we introduce UNeMo, a novel framework designed for the collaborative optimization of visual state reasoning and navigational decision-making. It introduces a Multimodal World Model (MWM) that takes visual features, language instructions, and navigational actions as inputs to jointly predict subsequent visual states, enabling cross-modal reasoning. Via a Hierarchical Prediction-Feedback (HPN) mechanism, MWM collaborates with navigation policies: the first layer generates actions using current vision-and-language features; MWM then infers post-action visual states to guide the second layer's fine-grained decisions. This forms a dynamic bidirectional promotion mechanism where MWM reasoning optimizes navigation policies, while policy decisions feedback to improve MWM's reasoning accuracy. Experiments on R2R and REVERIE datasets show UNeMo outperforms state-of-the-art methods by 2.1% and 0.7% in navigation accuracy for unseen scenes, validating its effectiveness. |
|
| User-Feedback-Driven Adaptation for Vision-and-Language Navigation | 2026-02-04 | ShowReal-world deployment of Vision-and-Language Navigation (VLN) agents is constrained by the scarcity of reliable supervision after offline training. While recent adaptation methods attempt to mitigate distribution shifts via environment-driven self-supervision (e.g., entropy minimization), these signals are often noisy and can cause the agent to amplify its own mistakes during long-horizon sequential decision-making. In this paper, we propose a paradigm shift that positions user feedback, specifically episode-level success confirmations and goal-level corrections, as a primary and general-purpose supervision signal for VLN. Unlike internal confidence scores, user feedback is intent-aligned and in-situ consistent, directly correcting the agent's decoupling from user instructions. To effectively leverage this supervision, we introduce a user-feedback-driven learning framework featuring a topology-aware trajectory construction pipeline. This mechanism lifts sparse, goal-level corrections into dense path-level supervision by generating feasible paths on the agent's incrementally built topological graph, enabling sample-efficient imitation learning without requiring step-by-step human demonstrations. Furthermore, we develop a persistent memory bank mechanism for warm-start initialization, supporting the reuse of previously acquired topology and cached representations across navigation sessions. Extensive experiments on the GSA-R2R benchmark demonstrate that our approach transforms sparse interaction into robust supervision, consistently outperforming environment-driven baselines while exhibiting strong adaptability across diverse instruction styles. |
|
| DV-VLN: Dual Verification for Reliable LLM-Based Vision-and-Language Navigation | 2026-01-26 | ShowVision-and-Language Navigation (VLN) requires an embodied agent to navigate in a complex 3D environment according to natural language instructions. Recent progress in large language models (LLMs) has enabled language-driven navigation with improved interpretability. However, most LLM-based agents still rely on single-shot action decisions, where the model must choose one option from noisy, textualized multi-perspective observations. Due to local mismatches and imperfect intermediate reasoning, such decisions can easily deviate from the correct path, leading to error accumulation and reduced reliability in unseen environments. In this paper, we propose DV-VLN, a new VLN framework that follows a generate-then-verify paradigm. DV-VLN first performs parameter-efficient in-domain adaptation of an open-source LLaMA-2 backbone to produce a structured navigational chain-of-thought, and then verifies candidate actions with two complementary channels: True-False Verification (TFV) and Masked-Entity Verification (MEV). DV-VLN selects actions by aggregating verification successes across multiple samples, yielding interpretable scores for reranking. Experiments on R2R, RxR (English subset), and REVERIE show that DV-VLN consistently improves over direct prediction and sampling-only baselines, achieving competitive performance among language-only VLN agents and promising results compared with several cross-modal systems.Code is available at https://github.com/PlumJun/DV-VLN. |
|
| VL-LN Bench: Towards Long-horizon Goal-oriented Navigation with Active Dialogs | 2026-01-23 | ShowIn most existing embodied navigation tasks, instructions are well-defined and unambiguous, such as instruction following and object searching. Under this idealized setting, agents are required solely to produce effective navigation outputs conditioned on vision and language inputs. However, real-world navigation instructions are often vague and ambiguous, requiring the agent to resolve uncertainty and infer user intent through active dialog. To address this gap, we propose Interactive Instance Goal Navigation (IIGN), a task that requires agents not only to generate navigation actions but also to produce language outputs via active dialog, thereby aligning more closely with practical settings. IIGN extends Instance Goal Navigation (IGN) by allowing agents to freely consult an oracle in natural language while navigating. Building on this task, we present the Vision Language-Language Navigation (VL-LN) benchmark, which provides a large-scale, automatically generated dataset and a comprehensive evaluation protocol for training and assessing dialog-enabled navigation models. VL-LN comprises over 41k long-horizon dialog-augmented trajectories for training and an automatic evaluation protocol with an oracle capable of responding to agent queries. Using this benchmark, we train a navigation model equipped with dialog capabilities and show that it achieves significant improvements over the baselines. Extensive experiments and analyses further demonstrate the effectiveness and reliability of VL-LN for advancing research on dialog-enabled embodied navigation. Code and dataset: https://0309hws.github.io/VL-LN.github.io/ |
|
| CLASH: Collaborative Large-Small Hierarchical Framework for Continuous Vision-and-Language Navigation | 2026-01-23 | ShowVision-and-Language Navigation (VLN) requires robots to follow natural language instructions and navigate complex environments without prior maps. While recent vision-language large models demonstrate strong reasoning abilities, they often underperform task-specific panoramic small models in VLN tasks. To address this, we propose CLASH (Collaborative Large-Small Hierarchy), a VLN-CE framework that integrates a reactive small-model planner (RSMP) with a reflective large-model reasoner (RLMR). RSMP adopts a causal-learning-based dual-branch architecture to enhance generalization, while RLMR leverages panoramic visual prompting with chain-of-thought reasoning to support interpretable spatial understanding and navigation. We further introduce an uncertainty-aware collaboration mechanism (UCM) that adaptively fuses decisions from both models. For obstacle avoidance, in simulation, we replace the rule-based controller with a fully learnable point-goal policy, and in real-world deployment, we design a LiDAR-based clustering module for generating navigable waypoints and pair it with an online SLAM-based local controller. CLASH achieves state-of-the-art (SoTA) results (ranking 1-st) on the VLN-CE leaderboard, significantly improving SR and SPL on the test-unseen set over the previous SoTA methods. Real-world experiments demonstrate CLASH's strong robustness, validating its effectiveness in both simulation and deployment scenarios. |
|
| FantasyVLN: Unified Multimodal Chain-of-Thought Reasoning for Vision-Language Navigation | 2026-01-23 | ShowAchieving human-level performance in Vision-and-Language Navigation (VLN) requires an embodied agent to jointly understand multimodal instructions and visual-spatial context while reasoning over long action sequences. Recent works, such as NavCoT and NavGPT-2, demonstrate the potential of Chain-of-Thought (CoT) reasoning for improving interpretability and long-horizon planning. Moreover, multimodal extensions like OctoNav-R1 and CoT-VLA further validate CoT as a promising pathway toward human-like navigation reasoning. However, existing approaches face critical drawbacks: purely textual CoTs lack spatial grounding and easily overfit to sparse annotated reasoning steps, while multimodal CoTs incur severe token inflation by generating imagined visual observations, making real-time navigation impractical. In this work, we propose FantasyVLN, a unified implicit reasoning framework that preserves the benefits of CoT reasoning without explicit token overhead. Specifically, imagined visual tokens are encoded into a compact latent space using a pretrained Visual AutoRegressor (VAR) during CoT reasoning training, and the model jointly learns from textual, visual, and multimodal CoT modes under a unified multi-CoT strategy. At inference, our model performs direct instruction-to-action mapping while still enjoying reasoning-aware representations. Extensive experiments on LH-VLN show that our approach achieves reasoning-aware yet real-time navigation, improving success rates and efficiency while reducing inference latency by an order of magnitude compared to explicit CoT methods. |
|
| Spatial-VLN: Zero-Shot Vision-and-Language Navigation With Explicit Spatial Perception and Exploration | 2026-01-19 | ShowZero-shot Vision-and-Language Navigation (VLN) agents leveraging Large Language Models (LLMs) excel in generalization but suffer from insufficient spatial perception. Focusing on complex continuous environments, we categorize key perceptual bottlenecks into three spatial challenges: door interaction,multi-room navigation, and ambiguous instruction execution, where existing methods consistently suffer high failure rates. We present Spatial-VLN, a perception-guided exploration framework designed to overcome these challenges. The framework consists of two main modules. The Spatial Perception Enhancement (SPE) module integrates panoramic filtering with specialized door and region experts to produce spatially coherent, cross-view consistent perceptual representations. Building on this foundation, our Explored Multi-expert Reasoning (EMR) module uses parallel LLM experts to address waypoint-level semantics and region-level spatial transitions. When discrepancies arise between expert predictions, a query-and-explore mechanism is activated, prompting the agent to actively probe critical areas and resolve perceptual ambiguities. Experiments on VLN-CE demonstrate that Spatial VLN achieves state-of-the-art performance using only low-cost LLMs. Furthermore, to validate real-world applicability, we introduce a value-based waypoint sampling strategy that effectively bridges the Sim2Real gap. Extensive real-world evaluations confirm that our framework delivers superior generalization and robustness in complex environments. Our codes and videos are available at https://yueluhhxx.github.io/Spatial-VLN-web/. |
|
| GROKE: Vision-Free Navigation Instruction Evaluation via Graph Reasoning on OpenStreetMap | 2026-01-12 | ShowThe evaluation of navigation instructions remains a persistent challenge in Vision-and-Language Navigation (VLN) research. Traditional reference-based metrics such as BLEU and ROUGE fail to capture the functional utility of spatial directives, specifically whether an instruction successfully guides a navigator to the intended destination. Although existing VLN agents could serve as evaluators, their reliance on high-fidelity visual simulators introduces licensing constraints and computational costs, and perception errors further confound linguistic quality assessment. This paper introduces GROKE(Graph-based Reasoning over OSM Knowledge for instruction Evaluation), a vision-free training-free hierarchical LLM-based framework for evaluating navigation instructions using OpenStreetMap data. Through systematic ablation studies, we demonstrate that structured JSON and textual formats for spatial information substantially outperform grid-based and visual graph representations. Our hierarchical architecture combines sub-instruction planning with topological graph navigation, reducing navigation error by 68.5% compared to heuristic and sampling baselines on the Map2Seq dataset. The agent's execution success, trajectory fidelity, and decision patterns serve as proxy metrics for functional navigability given OSM-visible landmarks and topology, establishing a scalable and interpretable evaluation paradigm without visual dependencies. Code and data are available at https://anonymous.4open.science/r/groke. |
Under...Under Review for ACL 2026 |
| SpatialNav: Leveraging Spatial Scene Graphs for Zero-Shot Vision-and-Language Navigation | 2026-01-11 | ShowAlthough learning-based vision-and-language navigation (VLN) agents can learn spatial knowledge implicitly from large-scale training data, zero-shot VLN agents lack this process, relying primarily on local observations for navigation, which leads to inefficient exploration and a significant performance gap. To deal with the problem, we consider a zero-shot VLN setting that agents are allowed to fully explore the environment before task execution. Then, we construct the Spatial Scene Graph (SSG) to explicitly capture global spatial structure and semantics in the explored environment. Based on the SSG, we introduce SpatialNav, a zero-shot VLN agent that integrates an agent-centric spatial map, a compass-aligned visual representation, and a remote object localization strategy for efficient navigation. Comprehensive experiments in both discrete and continuous environments demonstrate that SpatialNav significantly outperforms existing zero-shot agents and clearly narrows the gap with state-of-the-art learning-based methods. Such results highlight the importance of global spatial representations for generalizable navigation. |
11 pa...11 pages, 4 figures, 6 tables |
| SeqWalker: Sequential-Horizon Vision-and-Language Navigation with Hierarchical Planning | 2026-01-08 | ShowSequential-Horizon Vision-and-Language Navigation (SH-VLN) presents a challenging scenario where agents should sequentially execute multi-task navigation guided by complex, long-horizon language instructions. Current vision-and-language navigation models exhibit significant performance degradation with such multi-task instructions, as information overload impairs the agent's ability to attend to observationally relevant details. To address this problem, we propose SeqWalker, a navigation model built on a hierarchical planning framework. Our SeqWalker features: i) A High-Level Planner that dynamically selects global instructions into contextually relevant sub-instructions based on the agent's current visual observations, thus reducing cognitive load; ii) A Low-Level Planner incorporating an Exploration-Verification strategy that leverages the inherent logical structure of instructions for trajectory error correction. To evaluate SH-VLN performance, we also extend the IVLN dataset and establish a new benchmark. Extensive experiments are performed to demonstrate the superiority of the proposed SeqWalker. |
|
| AirNav: A Large-Scale Real-World UAV Vision-and-Language Navigation Dataset with Natural and Diverse Instructions | 2026-01-07 | ShowExisting Unmanned Aerial Vehicle (UAV) Vision-Language Navigation (VLN) datasets face issues such as dependence on virtual environments, lack of naturalness in instructions, and limited scale. To address these challenges, we propose AirNav, a large-scale UAV VLN benchmark constructed from real urban aerial data, rather than synthetic environments, with natural and diverse instructions. Additionally, we introduce the AirVLN-R1, which combines Supervised Fine-Tuning and Reinforcement Fine-Tuning to enhance performance and generalization. The feasibility of the model is preliminarily evaluated through real-world tests. Our dataset and code are publicly available. |
|
| VLN-MME: Diagnosing MLLMs as Language-guided Visual Navigation agents | 2026-01-06 | ShowMultimodal Large Language Models (MLLMs) have demonstrated remarkable capabilities across a wide range of vision-language tasks. However, their performance as embodied agents, which requires multi-round dialogue spatial reasoning and sequential action prediction, needs further exploration. Our work investigates this potential in the context of Vision-and-Language Navigation (VLN) by introducing a unified and extensible evaluation framework to probe MLLMs as zero-shot agents by bridging traditional navigation datasets into a standardized benchmark, named VLN-MME. We simplify the evaluation with a highly modular and accessible design. This flexibility streamlines experiments, enabling structured comparisons and component-level ablations across diverse MLLM architectures, agent designs, and navigation tasks. Crucially, enabled by our framework, we observe that enhancing our baseline agent with Chain-of-Thought (CoT) reasoning and self-reflection leads to an unexpected performance decrease. This suggests MLLMs exhibit poor context awareness in embodied navigation tasks; although they can follow instructions and structure their output, their 3D spatial reasoning fidelity is low. VLN-MME lays the groundwork for systematic evaluation of general-purpose MLLMs in embodied navigation settings and reveals limitations in their sequential decision-making capabilities. We believe these findings offer crucial guidance for MLLM post-training as embodied agents. |
|
| MDE-AgriVLN: Agricultural Vision-and-Language Navigation with Monocular Depth Estimation | 2026-01-01 | ShowAgricultural robots are serving as powerful assistants across a wide range of agricultural tasks, nevertheless, still heavily relying on manual operations or railway systems for movement. The AgriVLN method and the A2A benchmark pioneeringly extended Vision-and-Language Navigation (VLN) to the agricultural domain, enabling a robot to navigate to a target position following a natural language instruction. Unlike human binocular vision, most agricultural robots are only given a single camera for monocular vision, which results in limited spatial perception. To bridge this gap, we present the method of Agricultural Vision-and-Language Navigation with Monocular Depth Estimation (MDE-AgriVLN), in which we propose the MDE module generating depth features from RGB images, to assist the decision-maker on multimodal reasoning. When evaluated on the A2A benchmark, our MDE-AgriVLN method successfully increases Success Rate from 0.23 to 0.32 and decreases Navigation Error from 4.43m to 4.08m, demonstrating the state-of-the-art performance in the agricultural VLN domain. Code: https://github.com/AlexTraveling/MDE-AgriVLN. |
|
| LongFly: Long-Horizon UAV Vision-and-Language Navigation with Spatiotemporal Context Integration | 2025-12-26 | ShowUnmanned aerial vehicles (UAVs) are crucial tools for post-disaster search and rescue, facing challenges such as high information density, rapid changes in viewpoint, and dynamic structures, especially in long-horizon navigation. However, current UAV vision-and-language navigation(VLN) methods struggle to model long-horizon spatiotemporal context in complex environments, resulting in inaccurate semantic alignment and unstable path planning. To this end, we propose LongFly, a spatiotemporal context modeling framework for long-horizon UAV VLN. LongFly proposes a history-aware spatiotemporal modeling strategy that transforms fragmented and redundant historical data into structured, compact, and expressive representations. First, we propose the slot-based historical image compression module, which dynamically distills multi-view historical observations into fixed-length contextual representations. Then, the spatiotemporal trajectory encoding module is introduced to capture the temporal dynamics and spatial structure of UAV trajectories. Finally, to integrate existing spatiotemporal context with current observations, we design the prompt-guided multimodal integration module to support time-based reasoning and robust waypoint prediction. Experimental results demonstrate that LongFly outperforms state-of-the-art UAV VLN baselines by 7.89% in success rate and 6.33% in success weighted by path length, consistently across both seen and unseen environments. |
|
| Fine-Grained Instruction-Guided Graph Reasoning for Vision-and-Language Navigation | 2025-12-23 | ShowVision-and-Language Navigation (VLN) requires an embodied agent to traverse complex environments by following natural language instructions, demanding accurate alignment between visual observations and linguistic guidance. Despite recent progress, existing methods typically encode visual and directional cues in a coupled manner, and process instructions without explicitly extracting navigation-critical semantics, which often leads to imprecise spatial reasoning and suboptimal cross-modal alignment. To address these challenges, we propose a fine-grained instruction-guided graph reasoning framework (OIKG) that enhances both spatial representation and instruction understanding during navigation. Specifically, an observation-graph interaction mechanism is introduced to disentangle angular and visual cues while strengthening directed edge representations through geometric embedding, enabling more reliable spatial reasoning within the navigation graph. In addition, a fine-grained instruction guidance module is designed to explicitly extract and leverage location-specific and object-centric information from language instructions, facilitating more precise cross-modal alignment between linguistic semantics and navigable trajectories. By jointly integrating structured graph reasoning with instruction-critical semantic cues, the proposed approach significantly improves the agent's ability to follow complex navigation instructions. Extensive experiments on the R2R and RxR benchmarks demonstrate that our method consistently achieves state-of-the-art performance across multiple evaluation metrics, validating the effectiveness of fine-grained instruction-guided graph reasoning for vision-and-language navigation. |
10 pages, 4 figures |
| FreeAskWorld: An Interactive and Closed-Loop Simulator for Human-Centric Embodied AI | 2025-12-20 | ShowAs embodied intelligence emerges as a core frontier in artificial intelligence research, simulation platforms must evolve beyond low-level physical interactions to capture complex, human-centered social behaviors. We introduce FreeAskWorld, an interactive simulation framework that integrates large language models (LLMs) for high-level behavior planning and semantically grounded interaction, informed by theories of intention and social cognition. Our framework supports scalable, realistic human-agent simulations and includes a modular data generation pipeline tailored for diverse embodied tasks. To validate the framework, we extend the classic Vision-and-Language Navigation (VLN) task into a interaction enriched Direction Inquiry setting, wherein agents can actively seek and interpret navigational guidance. We present and publicly release FreeAskWorld, a large-scale benchmark dataset comprising reconstructed environments, six diverse task types, 16 core object categories, 63,429 annotated sample frames, and more than 17 hours of interaction data to support training and evaluation of embodied AI systems. We benchmark VLN models, and human participants under both open-loop and closed-loop settings. Experimental results demonstrate that models fine-tuned on FreeAskWorld outperform their original counterparts, achieving enhanced semantic understanding and interaction competency. These findings underscore the efficacy of socially grounded simulation frameworks in advancing embodied AI systems toward sophisticated high-level planning and more naturalistic human-agent interaction. Importantly, our work underscores that interaction itself serves as an additional information modality. |
9 pages, 4 figures |
| History-Enhanced Two-Stage Transformer for Aerial Vision-and-Language Navigation | 2025-12-17 | ShowAerial Vision-and-Language Navigation (AVLN) requires Unmanned Aerial Vehicle (UAV) agents to localize targets in large-scale urban environments based on linguistic instructions. While successful navigation demands both global environmental reasoning and local scene comprehension, existing UAV agents typically adopt mono-granularity frameworks that struggle to balance these two aspects. To address this limitation, this work proposes a History-Enhanced Two-Stage Transformer (HETT) framework, which integrates the two aspects through a coarse-to-fine navigation pipeline. Specifically, HETT first predicts coarse-grained target positions by fusing spatial landmarks and historical context, then refines actions via fine-grained visual analysis. In addition, a historical grid map is designed to dynamically aggregate visual features into a structured spatial memory, enhancing comprehensive scene awareness. Additionally, the CityNav dataset annotations are manually refined to enhance data quality. Experiments on the refined CityNav dataset show that HETT delivers significant performance gains, while extensive ablation studies further verify the effectiveness of each component. |
|
| D3D-VLP: Dynamic 3D Vision-Language-Planning Model for Embodied Grounding and Navigation | 2025-12-14 | ShowEmbodied agents face a critical dilemma that end-to-end models lack interpretability and explicit 3D reasoning, while modular systems ignore cross-component interdependencies and synergies. To bridge this gap, we propose the Dynamic 3D Vision-Language-Planning Model (D3D-VLP). Our model introduces two key innovations: 1) A Dynamic 3D Chain-of-Thought (3D CoT) that unifies planning, grounding, navigation, and question answering within a single 3D-VLM and CoT pipeline; 2) A Synergistic Learning from Fragmented Supervision (SLFS) strategy, which uses a masked autoregressive loss to learn from massive and partially-annotated hybrid data. This allows different CoT components to mutually reinforce and implicitly supervise each other. To this end, we construct a large-scale dataset with 10M hybrid samples from 5K real scans and 20K synthetic scenes that are compatible with online learning methods such as RL and DAgger. Our D3D-VLP achieves state-of-the-art results on multiple benchmarks, including Vision-and-Language Navigation (R2R-CE, REVERIE-CE, NavRAG-CE), Object-goal Navigation (HM3D-OVON), and Task-oriented Sequential Grounding and Navigation (SG3D). Real-world mobile manipulation experiments further validate the effectiveness. |
|
| Ground Slow, Move Fast: A Dual-System Foundation Model for Generalizable Vision-and-Language Navigation | 2025-12-09 | ShowWhile recent large vision-language models (VLMs) have improved generalization in vision-language navigation (VLN), existing methods typically rely on end-to-end pipelines that map vision-language inputs directly to short-horizon discrete actions. Such designs often produce fragmented motions, incur high latency, and struggle with real-world challenges like dynamic obstacle avoidance. We propose DualVLN, the first dual-system VLN foundation model that synergistically integrates high-level reasoning with low-level action execution. System 2, a VLM-based global planner, "grounds slowly" by predicting mid-term waypoint goals via image-grounded reasoning. System 1, a lightweight, multi-modal conditioning Diffusion Transformer policy, "moves fast" by leveraging both explicit pixel goals and latent features from System 2 to generate smooth and accurate trajectories. The dual-system design enables robust real-time control and adaptive local decision-making in complex, dynamic environments. By decoupling training, the VLM retains its generalization, while System 1 achieves interpretable and effective local navigation. DualVLN outperforms prior methods across all VLN benchmarks and real-world experiments demonstrate robust long-horizon planning and real-time adaptability in dynamic environments. |
|
| ST-Booster: An Iterative SpatioTemporal Perception Booster for Vision-and-Language Navigation in Continuous Environments | 2025-12-02 | ShowVision-and-Language Navigation in Continuous Environments (VLN-CE) requires agents to navigate unknown, continuous spaces based on natural language instructions. Compared to discrete settings, VLN-CE poses two core perception challenges. First, the absence of predefined observation points leads to heterogeneous visual memories and weakened global spatial correlations. Second, cumulative reconstruction errors in three-dimensional scenes introduce structural noise, impairing local feature perception. To address these challenges, this paper proposes ST-Booster, an iterative spatiotemporal booster that enhances navigation performance through multi-granularity perception and instruction-aware reasoning. ST-Booster consists of three key modules -- Hierarchical SpatioTemporal Encoding (HSTE), Multi-Granularity Aligned Fusion (MGAF), and ValueGuided Waypoint Generation (VGWG). HSTE encodes long-term global memory using topological graphs and captures shortterm local details via grid maps. MGAF aligns these dualmap representations with instructions through geometry-aware knowledge fusion. The resulting representations are iteratively refined through pretraining tasks. During reasoning, VGWG generates Guided Attention Heatmaps (GAHs) to explicitly model environment-instruction relevance and optimize waypoint selection. Extensive comparative experiments and performance analyses are conducted, demonstrating that ST-Booster outperforms existing state-of-the-art methods, particularly in complex, disturbance-prone environments. |
11 pages, 7 figures |
| Run, Ruminate, and Regulate: A Dual-process Thinking System for Vision-and-Language Navigation | 2025-11-18 | ShowVision-and-Language Navigation (VLN) requires an agent to dynamically explore complex 3D environments following human instructions. Recent research underscores the potential of harnessing large language models (LLMs) for VLN, given their commonsense knowledge and general reasoning capabilities. Despite their strengths, a substantial gap in task completion performance persists between LLM-based approaches and domain experts, as LLMs inherently struggle to comprehend real-world spatial correlations precisely. Additionally, introducing LLMs is accompanied with substantial computational cost and inference latency. To address these issues, we propose a novel dual-process thinking framework dubbed R3, integrating LLMs' generalization capabilities with VLN-specific expertise in a zero-shot manner. The framework comprises three core modules: Runner, Ruminator, and Regulator. The Runner is a lightweight transformer-based expert model that ensures efficient and accurate navigation under regular circumstances. The Ruminator employs a powerful multimodal LLM as the backbone and adopts chain-of-thought (CoT) prompting to elicit structured reasoning. The Regulator monitors the navigation progress and controls the appropriate thinking mode according to three criteria, integrating Runner and Ruminator harmoniously. Experimental results illustrate that R3 significantly outperforms other state-of-the-art methods, exceeding 3.28% and 3.30% in SPL and RGSPL respectively on the REVERIE benchmark. This pronounced enhancement highlights the effectiveness of our method in handling challenging VLN tasks. |
|
| Shedding Light on VLN Robustness: A Black-box Framework for Indoor Lighting-based Adversarial Attack | 2025-11-17 | ShowVision-and-Language Navigation (VLN) agents have made remarkable progress, but their robustness remains insufficiently studied. Existing adversarial evaluations often rely on perturbations that manifest as unusual textures rarely encountered in everyday indoor environments. Errors under such contrived conditions have limited practical relevance, as real-world agents are unlikely to encounter such artificial patterns. In this work, we focus on indoor lighting, an intrinsic yet largely overlooked scene attribute that strongly influences navigation. We propose Indoor Lighting-based Adversarial Attack (ILA), a black-box framework that manipulates global illumination to disrupt VLN agents. Motivated by typical household lighting usage, we design two attack modes: Static Indoor Lighting-based Attack (SILA), where the lighting intensity remains constant throughout an episode, and Dynamic Indoor Lighting-based Attack (DILA), where lights are switched on or off at critical moments to induce abrupt illumination changes. We evaluate ILA on two state-of-the-art VLN models across three navigation tasks. Results show that ILA significantly increases failure rates while reducing trajectory efficiency, revealing previously unrecognized vulnerabilities of VLN agents to realistic indoor lighting variations. |
|
| VISTAv2: World Imagination for Indoor Vision-and-Language Navigation | 2025-11-14 | ShowVision-and-Language Navigation (VLN) requires agents to follow language instructions while acting in continuous real-world spaces. Prior image imagination based VLN work shows benefits for discrete panoramas but lacks online, action-conditioned predictions and does not produce explicit planning values; moreover, many methods replace the planner with long-horizon objectives that are brittle and slow. To bridge this gap, we propose VISTAv2, a generative world model that rolls out egocentric future views conditioned on past observations, candidate action sequences, and instructions, and projects them into an online value map for planning. Unlike prior approaches, VISTAv2 does not replace the planner. The online value map is fused at score level with the base objective, providing reachability and risk-aware guidance. Concretely, we employ an action-aware Conditional Diffusion Transformer video predictor to synthesize short-horizon futures, align them with the natural language instruction via a vision-language scorer, and fuse multiple rollouts in a differentiable imagination-to-value head to output an imagined egocentric value map. For efficiency, rollouts occur in VAE latent space with a distilled sampler and sparse decoding, enabling inference on a single consumer GPU. Evaluated on MP3D and RoboTHOR, VISTAv2 improves over strong baselines, and ablations show that action-conditioned imagination, instruction-guided value fusion, and the online value-map planner are all critical, suggesting that VISTAv2 offers a practical and interpretable route to robust VLN. |
11 pages, 5 figures |
| Agent Journey Beyond RGB: Hierarchical Semantic-Spatial Representation Enrichment for Vision-and-Language Navigation | 2025-11-13 | ShowNavigating unseen environments from natural language instructions remains challenging for egocentric agents in Vision-and-Language Navigation (VLN). Humans naturally ground concrete semantic knowledge within spatial layouts during indoor navigation. Although prior work has introduced diverse environment representations to improve reasoning, auxiliary modalities are often naively concatenated with RGB features, which underutilizes each modality's distinct contribution. We propose a hierarchical Semantic Understanding and Spatial Awareness (SUSA) architecture to enable agents to perceive and ground environments at multiple scales. Specifically, the Textual Semantic Understanding (TSU) module supports local action prediction by generating view-level descriptions, capturing fine-grained semantics and narrowing the modality gap between instructions and environments. Complementarily, the Depth Enhanced Spatial Perception (DSP) module incrementally builds a trajectory-level depth exploration map, providing a coarse-grained representation of global spatial layout. Extensive experiments show that the hierarchical representation enrichment of SUSA significantly improves navigation performance over the baseline on discrete VLN benchmarks (REVERIE, R2R, and SOON) and generalizes better to the continuous R2R-CE benchmark. |
AAAI2...AAAI2026, I14 pages, 12 figures, 11 tables |
| A Survey on Improving Human Robot Collaboration through Vision-and-Language Navigation | 2025-11-06 | ShowVision-and-Language Navigation (VLN) is a multi-modal, cooperative task requiring agents to interpret human instructions, navigate 3D environments, and communicate effectively under ambiguity. This paper presents a comprehensive review of recent VLN advancements in robotics and outlines promising directions to improve multi-robot coordination. Despite progress, current models struggle with bidirectional communication, ambiguity resolution, and collaborative decision-making in the multi-agent systems. We review approximately 200 relevant articles to provide an in-depth understanding of the current landscape. Through this survey, we aim to provide a thorough resource that inspires further research at the intersection of VLN and robotics. We advocate that the future VLN systems should support proactive clarification, real-time feedback, and contextual reasoning through advanced natural language understanding (NLU) techniques. Additionally, decentralized decision-making frameworks with dynamic role assignment are essential for scalable, efficient multi-robot collaboration. These innovations can significantly enhance human-robot interaction (HRI) and enable real-world deployment in domains such as healthcare, logistics, and disaster response. |
|
| Fast-SmartWay: Panoramic-Free End-to-End Zero-Shot Vision-and-Language Navigation | 2025-11-02 | ShowRecent advances in Vision-and-Language Navigation in Continuous Environments (VLN-CE) have leveraged multimodal large language models (MLLMs) to achieve zero-shot navigation. However, existing methods often rely on panoramic observations and two-stage pipelines involving waypoint predictors, which introduce significant latency and limit real-world applicability. In this work, we propose Fast-SmartWay, an end-to-end zero-shot VLN-CE framework that eliminates the need for panoramic views and waypoint predictors. Our approach uses only three frontal RGB-D images combined with natural language instructions, enabling MLLMs to directly predict actions. To enhance decision robustness, we introduce an Uncertainty-Aware Reasoning module that integrates (i) a Disambiguation Module for avoiding local optima, and (ii) a Future-Past Bidirectional Reasoning mechanism for globally coherent planning. Experiments on both simulated and real-robot environments demonstrate that our method significantly reduces per-step latency while achieving competitive or superior performance compared to panoramic-view baselines. These results demonstrate the practicality and effectiveness of Fast-SmartWay for real-world zero-shot embodied navigation. |
|
| Continual Vision-and-Language Navigation | 2025-10-31 | ShowDeveloping Vision-and-Language Navigation (VLN) agents typically assumes a \textit{train-once-deploy-once} strategy, which is unrealistic as deployed agents continually encounter novel environments. To address this, we propose the Continual Vision-and-Language Navigation (CVLN) paradigm, where agents learn and adapt incrementally across multiple \textit{scene domains}. CVLN includes two setups: Initial-instruction based CVLN for instruction-following, and Dialogue-based CVLN for dialogue-guided navigation. We also introduce two simple yet effective baselines for sequential decision-making: Perplexity Replay (PerpR), which replays difficult episodes, and Episodic Self-Replay (ESR), which stores and revisits action logits during training. Experiments show that existing continual learning methods fall short for CVLN, while PerpR and ESR achieve better performance by efficiently utilizing replay memory. |
|
| STRIDER: Navigation via Instruction-Aligned Structural Decision Space Optimization | 2025-10-27 | ShowThe Zero-shot Vision-and-Language Navigation in Continuous Environments (VLN-CE) task requires agents to navigate previously unseen 3D environments using natural language instructions, without any scene-specific training. A critical challenge in this setting lies in ensuring agents' actions align with both spatial structure and task intent over long-horizon execution. Existing methods often fail to achieve robust navigation due to a lack of structured decision-making and insufficient integration of feedback from previous actions. To address these challenges, we propose STRIDER (Instruction-Aligned Structural Decision Space Optimization), a novel framework that systematically optimizes the agent's decision space by integrating spatial layout priors and dynamic task feedback. Our approach introduces two key innovations: 1) a Structured Waypoint Generator that constrains the action space through spatial structure, and 2) a Task-Alignment Regulator that adjusts behavior based on task progress, ensuring semantic alignment throughout navigation. Extensive experiments on the R2R-CE and RxR-CE benchmarks demonstrate that STRIDER significantly outperforms strong SOTA across key metrics; in particular, it improves Success Rate (SR) from 29% to 35%, a relative gain of 20.7%. Such results highlight the importance of spatially constrained decision-making and feedback-guided execution in improving navigation fidelity for zero-shot VLN-CE. |
|
| NavQ: Learning a Q-Model for Foresighted Vision-and-Language Navigation | 2025-10-18 | ShowIn this work we concentrate on the task of goal-oriented Vision-and-Language Navigation (VLN). Existing methods often make decisions based on historical information, overlooking the future implications and long-term outcomes of the actions. In contrast, we aim to develop a foresighted agent. Specifically, we draw upon Q-learning to train a Q-model using large-scale unlabeled trajectory data, in order to learn the general knowledge regarding the layout and object relations within indoor scenes. This model can generate a Q-feature, analogous to the Q-value in traditional Q-network, for each candidate action, which describes the potential future information that may be observed after taking the specific action. Subsequently, a cross-modal future encoder integrates the task-agnostic Q-feature with navigation instructions to produce a set of action scores reflecting future prospects. These scores, when combined with the original scores based on history, facilitate an A*-style searching strategy to effectively explore the regions that are more likely to lead to the destination. Extensive experiments conducted on widely used goal-oriented VLN datasets validate the effectiveness of the proposed method. |
ICCV 2025 |
| SUM-AgriVLN: Spatial Understanding Memory for Agricultural Vision-and-Language Navigation | 2025-10-16 | ShowAgricultural robots are emerging as powerful assistants across a wide range of agricultural tasks, nevertheless, still heavily rely on manual operation or fixed rail systems for movement. The AgriVLN method and the A2A benchmark pioneeringly extend Vision-and-Language Navigation (VLN) to the agricultural domain, enabling robots to navigate to the target positions following the natural language instructions. In practical agricultural scenarios, navigation instructions often repeatedly occur, yet AgriVLN treat each instruction as an independent episode, overlooking the potential of past experiences to provide spatial context for subsequent ones. To bridge this gap, we propose the method of Spatial Understanding Memory for Agricultural Vision-and-Language Navigation (SUM-AgriVLN), in which the SUM module employs spatial understanding and save spatial memory through 3D reconstruction and representation. When evaluated on the A2A benchmark, our SUM-AgriVLN effectively improves Success Rate from 0.47 to 0.54 with slight sacrifice on Navigation Error from 2.91m to 2.93m, demonstrating the state-of-the-art performance in the agricultural domain. Code: https://github.com/AlexTraveling/SUM-AgriVLN. |
|
| Mem4Nav: Boosting Vision-and-Language Navigation in Urban Environments with a Hierarchical Spatial-Cognition Long-Short Memory System | 2025-10-10 | ShowVision-and-Language Navigation (VLN) in large-scale urban environments requires embodied agents to ground linguistic instructions in complex scenes and recall relevant experiences over extended time horizons. Prior modular pipelines offer interpretability but lack unified memory, while end-to-end (M)LLM agents excel at fusing vision and language yet remain constrained by fixed context windows and implicit spatial reasoning. We introduce \textbf{Mem4Nav}, a hierarchical spatial-cognition long-short memory system that can augment any VLN backbone. Mem4Nav fuses a sparse octree for fine-grained voxel indexing with a semantic topology graph for high-level landmark connectivity, storing both in trainable memory tokens embedded via a reversible Transformer. Long-term memory (LTM) compresses and retains historical observations at both octree and graph nodes, while short-term memory (STM) caches recent multimodal entries in relative coordinates for real-time obstacle avoidance and local planning. At each step, STM retrieval sharply prunes dynamic context, and, when deeper history is needed, LTM tokens are decoded losslessly to reconstruct past embeddings. Evaluated on Touchdown and Map2Seq across three backbones (modular, state-of-the-art VLN with prompt-based LLM, and state-of-the-art VLN with strided-attention MLLM), Mem4Nav yields 7-13 pp gains in Task Completion, sufficient SPD reduction, and >10 pp nDTW improvement. Ablations confirm the indispensability of both the hierarchical map and dual memory modules. Our codes are open-sourced via https://github.com/tsinghua-fib-lab/Mem4Nav. |
The p...The paper is currently under investigation regarding concerns of potential academic misconduct. While the investigation is ongoing, the authors have voluntarily requested to withdraw the manuscript |
| HA-VLN 2.0: An Open Benchmark and Leaderboard for Human-Aware Navigation in Discrete and Continuous Environments with Dynamic Multi-Human Interactions | 2025-10-09 | ShowVision-and-Language Navigation (VLN) has been studied mainly in either discrete or continuous settings, with little attention to dynamic, crowded environments. We present HA-VLN 2.0, a unified benchmark introducing explicit social-awareness constraints. Our contributions are: (i) a standardized task and metrics capturing both goal accuracy and personal-space adherence; (ii) HAPS 2.0 dataset and simulators modeling multi-human interactions, outdoor contexts, and finer language-motion alignment; (iii) benchmarks on 16,844 socially grounded instructions, revealing sharp performance drops of leading agents under human dynamics and partial observability; and (iv) real-world robot experiments validating sim-to-real transfer, with an open leaderboard enabling transparent comparison. Results show that explicit social modeling improves navigation robustness and reduces collisions, underscoring the necessity of human-centric approaches. By releasing datasets, simulators, baselines, and protocols, HA-VLN 2.0 provides a strong foundation for safe, socially responsible navigation research. |
33 pa...33 pages, 20 figures, website: https://ha-vln-project.vercel.app/ |
| Breaking Down and Building Up: Mixture of Skill-Based Vision-and-Language Navigation Agents | 2025-10-01 | ShowVision-and-Language Navigation (VLN) poses significant challenges for agents to interpret natural language instructions and navigate complex 3D environments. While recent progress has been driven by large-scale pre-training and data augmentation, current methods still struggle to generalize to unseen scenarios, particularly when complex spatial and temporal reasoning is required. In this work, we propose SkillNav, a modular framework that introduces structured, skill-based reasoning into Transformer-based VLN agents. Our method decomposes navigation into a set of interpretable atomic skills (e.g., Vertical Movement, Area and Region Identification, Stop and Pause), each handled by a specialized agent. To support targeted skill training without manual data annotation, we construct a synthetic dataset pipeline that generates diverse, linguistically natural, skill-specific instruction-trajectory pairs. We then introduce a novel training-free Vision-Language Model (VLM)-based router, which dynamically selects the most suitable agent at each time step by aligning sub-goals with visual observations and historical actions. SkillNav obtains competitive results on commonly used benchmarks and establishes state-of-the-art generalization to the GSA-R2R, a benchmark with novel instruction styles and unseen environments. |
|
| Landmark-Guided Knowledge for Vision-and-Language Navigation | 2025-09-30 | ShowVision-and-language navigation is one of the core tasks in embodied intelligence, requiring an agent to autonomously navigate in an unfamiliar environment based on natural language instructions. However, existing methods often fail to match instructions with environmental information in complex scenarios, one reason being the lack of common-sense reasoning ability. This paper proposes a vision-and-language navigation method called Landmark-Guided Knowledge (LGK), which introduces an external knowledge base to assist navigation, addressing the misjudgment issues caused by insufficient common sense in traditional methods. Specifically, we first construct a knowledge base containing 630,000 language descriptions and use knowledge Matching to align environmental subviews with the knowledge base, extracting relevant descriptive knowledge. Next, we design a Knowledge-Guided by Landmark (KGL) mechanism, which guides the agent to focus on the most relevant parts of the knowledge by leveraging landmark information in the instructions, thereby reducing the data bias that may arise from incorporating external knowledge. Finally, we propose Knowledge-Guided Dynamic Augmentation (KGDA), which effectively integrates language, knowledge, vision, and historical information. Experimental results demonstrate that the LGK method outperforms existing state-of-the-art methods on the R2R and REVERIE vision-and-language navigation datasets, particularly in terms of navigation error, success rate, and path efficiency. |
Accep...Accepted for publication by International Conference on Intelligent Computing 2025 |
| Vision-and-Language Navigation with Analogical Textual Descriptions in LLMs | 2025-09-29 | ShowIntegrating large language models (LLMs) into embodied AI models is becoming increasingly prevalent. However, existing zero-shot LLM-based Vision-and-Language Navigation (VLN) agents either encode images as textual scene descriptions, potentially oversimplifying visual details, or process raw image inputs, which can fail to capture abstract semantics required for high-level reasoning. In this paper, we improve the navigation agent's contextual understanding by incorporating textual descriptions from multiple perspectives that facilitate analogical reasoning across images. By leveraging text-based analogical reasoning, the agent enhances its global scene understanding and spatial reasoning, leading to more accurate action decisions. We evaluate our approach on the R2R dataset, where our experiments demonstrate significant improvements in navigation performance. |
|
| See, Point, Fly: A Learning-Free VLM Framework for Universal Unmanned Aerial Navigation | 2025-09-26 | ShowWe present See, Point, Fly (SPF), a training-free aerial vision-and-language navigation (AVLN) framework built atop vision-language models (VLMs). SPF is capable of navigating to any goal based on any type of free-form instructions in any kind of environment. In contrast to existing VLM-based approaches that treat action prediction as a text generation task, our key insight is to consider action prediction for AVLN as a 2D spatial grounding task. SPF harnesses VLMs to decompose vague language instructions into iterative annotation of 2D waypoints on the input image. Along with the predicted traveling distance, SPF transforms predicted 2D waypoints into 3D displacement vectors as action commands for UAVs. Moreover, SPF also adaptively adjusts the traveling distance to facilitate more efficient navigation. Notably, SPF performs navigation in a closed-loop control manner, enabling UAVs to follow dynamic targets in dynamic environments. SPF sets a new state of the art in DRL simulation benchmark, outperforming the previous best method by an absolute margin of 63%. In extensive real-world evaluations, SPF outperforms strong baselines by a large margin. We also conduct comprehensive ablation studies to highlight the effectiveness of our design choice. Lastly, SPF shows remarkable generalization to different VLMs. Project page: https://spf-web.pages.dev |
CoRL ...CoRL 2025. Project page: https://spf-web.pages.dev |
| Rethinking the Embodied Gap in Vision-and-Language Navigation: A Holistic Study of Physical and Visual Disparities | 2025-09-26 | ShowRecent Vision-and-Language Navigation (VLN) advancements are promising, but their idealized assumptions about robot movement and control fail to reflect physically embodied deployment challenges. To bridge this gap, we introduce VLN-PE, a physically realistic VLN platform supporting humanoid, quadruped, and wheeled robots. For the first time, we systematically evaluate several ego-centric VLN methods in physical robotic settings across different technical pipelines, including classification models for single-step discrete action prediction, a diffusion model for dense waypoint prediction, and a train-free, map-based large language model (LLM) integrated with path planning. Our results reveal significant performance degradation due to limited robot observation space, environmental lighting variations, and physical challenges like collisions and falls. This also exposes locomotion constraints for legged robots in complex environments. VLN-PE is highly extensible, allowing seamless integration of new scenes beyond MP3D, thereby enabling more comprehensive VLN evaluation. Despite the weak generalization of current models in physical deployment, VLN-PE provides a new pathway for improving cross-embodiment's overall adaptability. We hope our findings and tools inspire the community to rethink VLN limitations and advance robust, practical VLN models. The code is available at https://crystalsixone.github.io/vln_pe.github.io/. |
Accep...Accepted by ICCV 2025 |
| Walk and Read Less: Improving the Efficiency of Vision-and-Language Navigation via Tuning-Free Multimodal Token Pruning | 2025-09-22 | ShowLarge models achieve strong performance on Vision-and-Language Navigation (VLN) tasks, but are costly to run in resource-limited environments. Token pruning offers appealing tradeoffs for efficiency with minimal performance loss by reducing model input size, but prior work overlooks VLN-specific challenges. For example, information loss from pruning can effectively increase computational cost due to longer walks. Thus, the inability to identify uninformative tokens undermines the supposed efficiency gains from pruning. To address this, we propose Navigation-Aware Pruning (NAP), which uses navigation-specific traits to simplify the pruning process by pre-filtering tokens into foreground and background. For example, image views are filtered based on whether the agent can navigate in that direction. We also extract navigation-relevant instructions using a Large Language Model. After filtering, we focus pruning on background tokens, minimizing information loss. To further help avoid increases in navigation length, we discourage backtracking by removing low-importance navigation nodes. Experiments on standard VLN benchmarks show NAP significantly outperforms prior work, preserving higher success rates while saving more than 50% FLOPS. |
Accep...Accepted to EMNLP 2025. Data and code to be released at https://github.com/wdqin/VLN-NAP |
| Embodied Navigation Foundation Model | 2025-09-16 | ShowNavigation is a fundamental capability in embodied AI, representing the intelligence required to perceive and interact within physical environments following language instructions. Despite significant progress in large Vision-Language Models (VLMs), which exhibit remarkable zero-shot performance on general vision-language tasks, their generalization ability in embodied navigation remains largely confined to narrow task settings and embodiment-specific architectures. In this work, we introduce a cross-embodiment and cross-task Navigation Foundation Model (NavFoM), trained on eight million navigation samples that encompass quadrupeds, drones, wheeled robots, and vehicles, and spanning diverse tasks such as vision-and-language navigation, object searching, target tracking, and autonomous driving. NavFoM employs a unified architecture that processes multimodal navigation inputs from varying camera configurations and navigation horizons. To accommodate diverse camera setups and temporal horizons, NavFoM incorporates identifier tokens that embed camera view information of embodiments and the temporal context of tasks. Furthermore, to meet the demands of real-world deployment, NavFoM controls all observation tokens using a dynamically adjusted sampling strategy under a limited token length budget. Extensive evaluations on public benchmarks demonstrate that our model achieves state-of-the-art or highly competitive performance across multiple navigation tasks and embodiments without requiring task-specific fine-tuning. Additional real-world experiments further confirm the strong generalization capability and practical applicability of our approach. |
Proje...Project Page: https://pku-epic.github.io/NavFoM-Web/ |
| ActiveVLN: Towards Active Exploration via Multi-Turn RL in Vision-and-Language Navigation | 2025-09-16 | ShowThe Vision-and-Language Navigation (VLN) task requires an agent to follow natural language instructions and navigate through complex environments. Existing MLLM-based VLN methods primarily rely on imitation learning (IL) and often use DAgger for post-training to mitigate covariate shift. While effective, these approaches incur substantial data collection and training costs. Reinforcement learning (RL) offers a promising alternative. However, prior VLN RL methods lack dynamic interaction with the environment and depend on expert trajectories for reward shaping, rather than engaging in open-ended active exploration. This restricts the agent's ability to discover diverse and plausible navigation routes. To address these limitations, we propose ActiveVLN, a VLN framework that explicitly enables active exploration through multi-turn RL. In the first stage, a small fraction of expert trajectories is used for IL to bootstrap the agent. In the second stage, the agent iteratively predicts and executes actions, automatically collects diverse trajectories, and optimizes multiple rollouts via the GRPO objective. To further improve RL efficiency, we introduce a dynamic early-stopping strategy to prune long-tail or likely failed trajectories, along with additional engineering optimizations. Experiments show that ActiveVLN achieves the largest performance gains over IL baselines compared to both DAgger-based and prior RL-based post-training methods, while reaching competitive performance with state-of-the-art approaches despite using a smaller model. Code and data will be released soon. |
|
| DreamNav: A Trajectory-Based Imaginative Framework for Zero-Shot Vision-and-Language Navigation | 2025-09-14 | ShowVision-and-Language Navigation in Continuous Environments (VLN-CE), which links language instructions to perception and control in the real world, is a core capability of embodied robots. Recently, large-scale pretrained foundation models have been leveraged as shared priors for perception, reasoning, and action, enabling zero-shot VLN without task-specific training. However, existing zero-shot VLN methods depend on costly perception and passive scene understanding, collapsing control to point-level choices. As a result, they are expensive to deploy, misaligned in action semantics, and short-sighted in planning. To address these issues, we present DreamNav that focuses on the following three aspects: (1) for reducing sensory cost, our EgoView Corrector aligns viewpoints and stabilizes egocentric perception; (2) instead of point-level actions, our Trajectory Predictor favors global trajectory-level planning to better align with instruction semantics; and (3) to enable anticipatory and long-horizon planning, we propose an Imagination Predictor to endow the agent with proactive thinking capability. On VLN-CE and real-world tests, DreamNav sets a new zero-shot state-of-the-art (SOTA), outperforming the strongest egocentric baseline with extra information by up to 7.49% and 18.15% in terms of SR and SPL metrics. To our knowledge, this is the first zero-shot VLN method to unify trajectory-level planning and active imagination while using only egocentric inputs. |
|
| GC-VLN: Instruction as Graph Constraints for Training-free Vision-and-Language Navigation | 2025-09-12 | ShowIn this paper, we propose a training-free framework for vision-and-language navigation (VLN). Existing zero-shot VLN methods are mainly designed for discrete environments or involve unsupervised training in continuous simulator environments, which makes it challenging to generalize and deploy them in real-world scenarios. To achieve a training-free framework in continuous environments, our framework formulates navigation guidance as graph constraint optimization by decomposing instructions into explicit spatial constraints. The constraint-driven paradigm decodes spatial semantics through constraint solving, enabling zero-shot adaptation to unseen environments. Specifically, we construct a spatial constraint library covering all types of spatial relationship mentioned in VLN instructions. The human instruction is decomposed into a directed acyclic graph, with waypoint nodes, object nodes and edges, which are used as queries to retrieve the library to build the graph constraints. The graph constraint optimization is solved by the constraint solver to determine the positions of waypoints, obtaining the robot's navigation path and final goal. To handle cases of no solution or multiple solutions, we construct a navigation tree and the backtracking mechanism. Extensive experiments on standard benchmarks demonstrate significant improvements in success rate and navigation efficiency compared to state-of-the-art zero-shot VLN methods. We further conduct real-world experiments to show that our framework can effectively generalize to new environments and instruction sets, paving the way for a more robust and autonomous navigation framework. |
Accep...Accepted to CoRL 2025. Project page: this https URL |
| MSNav: Zero-Shot Vision-and-Language Navigation with Dynamic Memory and LLM Spatial Reasoning | 2025-09-10 | ShowVision-and-Language Navigation (VLN) requires an agent to interpret natural language instructions and navigate complex environments. Current approaches often adopt a "black-box" paradigm, where a single Large Language Model (LLM) makes end-to-end decisions. However, it is plagued by critical vulnerabilities, including poor spatial reasoning, weak cross-modal grounding, and memory overload in long-horizon tasks. To systematically address these issues, we propose Memory Spatial Navigation(MSNav), a framework that fuses three modules into a synergistic architecture, which transforms fragile inference into a robust, integrated intelligence. MSNav integrates three modules: Memory Module, a dynamic map memory module that tackles memory overload through selective node pruning, enhancing long-range exploration; Spatial Module, a module for spatial reasoning and object relationship inference that improves endpoint recognition; and Decision Module, a module using LLM-based path planning to execute robust actions. Powering Spatial Module, we also introduce an Instruction-Object-Space (I-O-S) dataset and fine-tune the Qwen3-4B model into Qwen-Spatial (Qwen-Sp), which outperforms leading commercial LLMs in object list extraction, achieving higher F1 and NDCG scores on the I-O-S test set. Extensive experiments on the Room-to-Room (R2R) and REVERIE datasets demonstrate MSNav's state-of-the-art performance with significant improvements in Success Rate (SR) and Success weighted by Path Length (SPL). |
9 pages, 4 figures |
| UAV-ON: A Benchmark for Open-World Object Goal Navigation with Aerial Agents | 2025-08-22 | ShowAerial navigation is a fundamental yet underexplored capability in embodied intelligence, enabling agents to operate in large-scale, unstructured environments where traditional navigation paradigms fall short. However, most existing research follows the Vision-and-Language Navigation (VLN) paradigm, which heavily depends on sequential linguistic instructions, limiting its scalability and autonomy. To address this gap, we introduce UAV-ON, a benchmark for large-scale Object Goal Navigation (ObjectNav) by aerial agents in open-world environments, where agents operate based on high-level semantic goals without relying on detailed instructional guidance as in VLN. UAV-ON comprises 14 high-fidelity Unreal Engine environments with diverse semantic regions and complex spatial layouts, covering urban, natural, and mixed-use settings. It defines 1270 annotated target objects, each characterized by an instance-level instruction that encodes category, physical footprint, and visual descriptors, allowing grounded reasoning. These instructions serve as semantic goals, introducing realistic ambiguity and complex reasoning challenges for aerial agents. To evaluate the benchmark, we implement several baseline methods, including Aerial ObjectNav Agent (AOA), a modular policy that integrates instruction semantics with egocentric observations for long-horizon, goal-directed exploration. Empirical results show that all baselines struggle in this setting, highlighting the compounded challenges of aerial navigation and semantic goal grounding. UAV-ON aims to advance research on scalable UAV autonomy driven by semantic goal descriptions in complex real-world environments. |
Accep...Accepted to ACM MM Dataset Track 2025 |
| AeroDuo: Aerial Duo for UAV-based Vision and Language Navigation | 2025-08-21 | ShowAerial Vision-and-Language Navigation (VLN) is an emerging task that enables Unmanned Aerial Vehicles (UAVs) to navigate outdoor environments using natural language instructions and visual cues. However, due to the extended trajectories and complex maneuverability of UAVs, achieving reliable UAV-VLN performance is challenging and often requires human intervention or overly detailed instructions. To harness the advantages of UAVs' high mobility, which could provide multi-grained perspectives, while maintaining a manageable motion space for learning, we introduce a novel task called Dual-Altitude UAV Collaborative VLN (DuAl-VLN). In this task, two UAVs operate at distinct altitudes: a high-altitude UAV responsible for broad environmental reasoning, and a low-altitude UAV tasked with precise navigation. To support the training and evaluation of the DuAl-VLN, we construct the HaL-13k, a dataset comprising 13,838 collaborative high-low UAV demonstration trajectories, each paired with target-oriented language instructions. This dataset includes both unseen maps and an unseen object validation set to systematically evaluate the model's generalization capabilities across novel environments and unfamiliar targets. To consolidate their complementary strengths, we propose a dual-UAV collaborative VLN framework, AeroDuo, where the high-altitude UAV integrates a multimodal large language model (Pilot-LLM) for target reasoning, while the low-altitude UAV employs a lightweight multi-stage policy for navigation and target grounding. The two UAVs work collaboratively and only exchange minimal coordinate information to ensure efficiency. |
Accep...Accepted by ACM MM 2025 |
| CorrectNav: Self-Correction Flywheel Empowers Vision-Language-Action Navigation Model | 2025-08-14 | ShowExisting vision-and-language navigation models often deviate from the correct trajectory when executing instructions. However, these models lack effective error correction capability, hindering their recovery from errors. To address this challenge, we propose Self-correction Flywheel, a novel post-training paradigm. Instead of considering the model's error trajectories on the training set as a drawback, our paradigm emphasizes their significance as a valuable data source. We have developed a method to identify deviations in these error trajectories and devised innovative techniques to automatically generate self-correction data for perception and action. These self-correction data serve as fuel to power the model's continued training. The brilliance of our paradigm is revealed when we re-evaluate the model on the training set, uncovering new error trajectories. At this time, the self-correction flywheel begins to spin. Through multiple flywheel iterations, we progressively enhance our monocular RGB-based VLA navigation model CorrectNav. Experiments on R2R-CE and RxR-CE benchmarks show CorrectNav achieves new state-of-the-art success rates of 65.1% and 69.3%, surpassing prior best VLA navigation models by 8.2% and 16.4%. Real robot tests in various indoor and outdoor environments demonstrate \method's superior capability of error correction, dynamic obstacle avoidance, and long instruction following. |
|
| Harnessing Input-Adaptive Inference for Efficient VLN | 2025-08-12 | ShowAn emerging paradigm in vision-and-language navigation (VLN) is the use of history-aware multi-modal transformer models. Given a language instruction, these models process observation and navigation history to predict the most appropriate action for an agent. While they have significantly improved performance, the scale of these models can be a bottleneck in practical settings with limited computational resources. In this work, we propose a novel input-adaptive navigation method to enhance VLN model efficiency. We first show that existing input-adaptive mechanisms fail to reduce computations without substantial performance degradation. To address this, we introduce three adaptive algorithms, each deployed at a different level: (1) To improve spatial efficiency, we selectively process panoramic views at each observation of an agent. (2) To improve intra-model efficiency, we propose importance-based adaptive thresholding for the early-exit methods. (3) To improve temporal efficiency, we implement a caching mechanism that prevents reprocessing of views previously seen by the agent. In evaluations on seven VLN benchmarks, we demonstrate over a 2$\times$ reduction in computation across three off-the-shelf agents in both standard and continuous environments. Our code is publicly available at https://github.com/secure-ai-systems-group/adaptive-vision-and-language-navigation. |
Accep...Accepted to ICCV 2025 [Poster] |
| Exploring Spatial Representation to Enhance LLM Reasoning in Aerial Vision-Language Navigation | 2025-08-11 | ShowAerial Vision-and-Language Navigation (VLN) is a novel task enabling Unmanned Aerial Vehicles (UAVs) to navigate in outdoor environments through natural language instructions and visual cues. However, it remains challenging due to the complex spatial relationships in aerial scenes.In this paper, we propose a training-free, zero-shot framework for aerial VLN tasks, where the large language model (LLM) is leveraged as the agent for action prediction. Specifically, we develop a novel Semantic-Topo-Metric Representation (STMR) to enhance the spatial reasoning capabilities of LLMs. This is achieved by extracting and projecting instruction-related semantic masks onto a top-down map, which presents spatial and topological information about surrounding landmarks and grows during the navigation process. At each step, a local map centered at the UAV is extracted from the growing top-down map, and transformed into a ma trix representation with distance metrics, serving as the text prompt to LLM for action prediction in response to the given instruction. Experiments conducted in real and simulation environments have proved the effectiveness and robustness of our method, achieving absolute success rate improvements of 26.8% and 5.8% over current state-of-the-art methods on simple and complex navigation tasks, respectively. The dataset and code will be released soon. |
|
| AgriVLN: Vision-and-Language Navigation for Agricultural Robots | 2025-08-10 | ShowAgricultural robots have emerged as powerful members in agricultural tasks, nevertheless, still heavily rely on manual operation or untransportable railway for movement, resulting in limited mobility and poor adaptability. Vision-and-Language Navigation (VLN) enables robots to navigate to the target destinations following natural language instructions, demonstrating strong performance on several domains. However, none of the existing benchmarks or methods is specifically designed for agricultural scenes. To bridge this gap, we propose Agriculture to Agriculture (A2A) benchmark, containing 1,560 episodes across six diverse agricultural scenes, in which all realistic RGB videos are captured by front-facing camera on a quadruped robot at a height of 0.38 meters, aligning with the practical deployment conditions. Meanwhile, we propose Vision-and-Language Navigation for Agricultural Robots (AgriVLN) baseline based on Vision-Language Model (VLM) prompted with carefully crafted templates, which can understand both given instructions and agricultural environments to generate appropriate low-level actions for robot control. When evaluated on A2A, AgriVLN performs well on short instructions but struggles with long instructions, because it often fails to track which part of the instruction is currently being executed. To address this, we further propose Subtask List (STL) instruction decomposition module and integrate it into AgriVLN, improving Success Rate (SR) from 0.33 to 0.47. We additionally compare AgriVLN with several existing VLN methods, demonstrating the state-of-the-art performance in the agricultural domain. |
|
| Following Route Instructions using Large Vision-Language Models: A Comparison between Low-level and Panoramic Action Spaces | 2025-08-04 | ShowVision-and-Language Navigation (VLN) refers to the task of enabling autonomous robots to navigate unfamiliar environments by following natural language instructions. While recent Large Vision-Language Models (LVLMs) have shown promise in this task, most current VLM systems rely on models specifically designed and optimized for navigation, leaving the potential of off-the-shelf LVLMs underexplored. Furthermore, while older VLN approaches used low-level action spaces with egocentric views and atomic actions (such as "turn left" or "move forward"), newer models tend to favor panoramic action spaces with discrete navigable viewpoints. This paper investigates (1) whether off-the-shelf LVLMs (fine-tuned without architectural modifications or simulator-based training) can effectively support VLN tasks and (2) whether such models can support both low-level and panoramic action paradigms. To this end, we fine-tune the open-source model Qwen2.5-VL-3B-Instruct on the Room-to-Room (R2R) dataset and evaluate its empirical performance across both low-level and panoramic action spaces. The best resulting model achieves a 41% success rate on the R2R test set, demonstrating that while off-the-shelf LVLMs can learn to perform Vision-and-Language Navigation, they still lag behind models specifically designed for this task. |
This ...This paper has been accepted to ICNSLP 2025 |
| CityNav: A Large-Scale Dataset for Real-World Aerial Navigation | 2025-08-02 | ShowVision-and-language navigation (VLN) aims to develop agents capable of navigating in realistic environments. While recent cross-modal training approaches have significantly improved navigation performance in both indoor and outdoor scenarios, aerial navigation over real-world cities remains underexplored primarily due to limited datasets and the difficulty of integrating visual and geographic information. To fill this gap, we introduce CityNav, the first large-scale real-world dataset for aerial VLN. Our dataset consists of 32,637 human demonstration trajectories, each paired with a natural language description, covering 4.65 km$^2$ across two real cities: Cambridge and Birmingham. In contrast to existing datasets composed of synthetic scenes such as AerialVLN, our dataset presents a unique challenge because agents must interpret spatial relationships between real-world landmarks and the navigation destination, making CityNav an essential benchmark for advancing aerial VLN. Furthermore, as an initial step toward addressing this challenge, we provide a methodology of creating geographic semantic maps that can be used as an auxiliary modality input during navigation. In our experiments, we compare performance of three representative aerial VLN agents (Seq2seq, CMA and AerialVLN models) and demonstrate that the semantic map representation significantly improves their navigation performance. |
ICCV2...ICCV2025. The first two authors are equally contributed. Project page: https://water-cookie.github.io/city-nav-proj/ |
| NavMorph: A Self-Evolving World Model for Vision-and-Language Navigation in Continuous Environments | 2025-07-22 | ShowVision-and-Language Navigation in Continuous Environments (VLN-CE) requires agents to execute sequential navigation actions in complex environments guided by natural language instructions. Current approaches often struggle with generalizing to novel environments and adapting to ongoing changes during navigation. Inspired by human cognition, we present NavMorph, a self-evolving world model framework that enhances environmental understanding and decision-making in VLN-CE tasks. NavMorph employs compact latent representations to model environmental dynamics, equipping agents with foresight for adaptive planning and policy refinement. By integrating a novel Contextual Evolution Memory, NavMorph leverages scene-contextual information to support effective navigation while maintaining online adaptability. Extensive experiments demonstrate that our method achieves notable performance improvements on popular VLN-CE benchmarks. Code is available at https://github.com/Feliciaxyao/NavMorph. |
Accep...Accepted by ICCV 2025 |
| MapNav: A Novel Memory Representation via Annotated Semantic Maps for Vision-and-Language Navigation | 2025-07-10 | ShowVision-and-language navigation (VLN) is a key task in Embodied AI, requiring agents to navigate diverse and unseen environments while following natural language instructions. Traditional approaches rely heavily on historical observations as spatio-temporal contexts for decision making, leading to significant storage and computational overhead. In this paper, we introduce MapNav, a novel end-to-end VLN model that leverages Annotated Semantic Map (ASM) to replace historical frames. Specifically, our approach constructs a top-down semantic map at the start of each episode and update it at each timestep, allowing for precise object mapping and structured navigation information. Then, we enhance this map with explicit textual labels for key regions, transforming abstract semantics into clear navigation cues and generate our ASM. MapNav agent using the constructed ASM as input, and use the powerful end-to-end capabilities of VLM to empower VLN. Extensive experiments demonstrate that MapNav achieves state-of-the-art (SOTA) performance in both simulated and real-world environments, validating the effectiveness of our method. Moreover, we will release our ASM generation source code and dataset to ensure reproducibility, contributing valuable resources to the field. We believe that our proposed MapNav can be used as a new memory representation method in VLN, paving the way for future research in this field. |
| Title | Date | Abstract | Comment |
|---|---|---|---|
| The Compression Gap: Why Discrete Tokenization Limits Vision-Language-Action Model Scaling | 2026-04-03 | ShowScaling Vision-Language-Action (VLA) models by upgrading the vision encoder is expected to improve downstream manipulation performance--as it does in vision-language modeling. We show that this expectation fails when actions are represented as discrete tokens, and explain why through an information-theoretic principle we call the Compression Gap: in any visuomotor pipeline, scaling behavior is governed by the location of the tightest information bottleneck. When actions are continuous (e.g., Diffusion Policy), the vision encoder is the binding constraint, and upgrading it directly improves performance. When actions are discretized through a fixed-capacity codebook (e.g., OAT), the codebook becomes the binding constraint, and encoder improvements cannot propagate past it--regardless of how rich the upstream representation is. We validate this principle on the LIBERO benchmark with three lines of evidence: a factorial experiment showing that encoder upgrades improve Diffusion Policy by over 21 percentage points while OAT gains are substantially attenuated across model scales; an encoder quality gradient across four encoders confirming that Diffusion Policy tracks encoder quality monotonically while OAT remains flat; and a codebook size experiment demonstrating that relaxing codebook capacity partially recovers encoder sensitivity, providing causal evidence for the bottleneck hypothesis. Our findings reveal that scaling in Physical AI requires identifying where information bottlenecks lie in the pipeline, rather than uniformly increasing model or data size. |
11 pages, 1 figure |
| Multi-View Video Diffusion Policy: A 3D Spatio-Temporal-Aware Video Action Model | 2026-04-03 | ShowRobotic manipulation requires understanding both the 3D spatial structure of the environment and its temporal evolution, yet most existing policies overlook one or both. They typically rely on 2D visual observations and backbones pretrained on static image--text pairs, resulting in high data requirements and limited understanding of environment dynamics. To address this, we introduce MV-VDP, a multi-view video diffusion policy that jointly models the 3D spatio-temporal state of the environment. The core idea is to simultaneously predict multi-view heatmap videos and RGB videos, which 1) align the representation format of video pretraining with action finetuning, and 2) specify not only what actions the robot should take, but also how the environment is expected to evolve in response to those actions. Extensive experiments show that MV-VDP enables data-efficient, robust, generalizable, and interpretable manipulation. With only ten demonstration trajectories and without additional pretraining, MV-VDP successfully performs complex real-world tasks, demonstrates strong robustness across a range of model hyperparameters, generalizes to out-of-distribution settings, and predicts realistic future videos. Experiments on Meta-World and real-world robotic platforms demonstrate that MV-VDP consistently outperforms video-prediction--based, 3D-based, and vision--language--action models, establishing a new state of the art in data-efficient multi-task manipulation. |
Proje...Project Website: https://lpy1219.github.io/MV-VDP-Web/ |
| Open-Loop Planning, Closed-Loop Verification: Speculative Verification for VLA | 2026-04-03 | ShowVision-Language-Action (VLA) models, as large foundation models for embodied control, have shown strong performance in manipulation tasks. However, their performance comes at high inference cost. To improve efficiency, recent methods adopt action chunking, which predicts a sequence of future actions for open-loop execution. Although effective for reducing computation, open-loop execution is sensitive to environmental changes and prone to error accumulation due to the lack of close-loop feedback. To address this limitation, we propose Speculative Verification for VLA Control (SV-VLA), a framework that combines efficient open-loop long-horizon planning with lightweight closed-loop online verification. Specifically, SV-VLA uses a heavy VLA as a low-frequency macro-planner to generate an action chunk together with a planning context, while a lightweight verifier continuously monitors execution based on the latest observations. Conditioned on both the current observation and the planning context, the verifier compares the planned action against a closed-loop reference action and triggers replanning only when necessary. Experiments demonstrate that SV-VLA combines the efficiency of chunked prediction with the robustness of closed-loop control, enabling efficient and reliable VLA-based control in dynamic environments. Code is available: https://github.com/edsad122/SV-VLA. |
Under Review |
| Look, Zoom, Understand: The Robotic Eyeball for Embodied Perception | 2026-04-03 | ShowIn embodied AI, visual perception should be active rather than passive: the system must decide where to look and at what scale to sense to acquire maximally informative data under pixel and spatial budget constraints. Existing vision models coupled with fixed RGB-D cameras fundamentally fail to reconcile wide-area coverage with fine-grained detail acquisition, severely limiting their efficacy in open-world robotic applications. We study the task of language-guided active visual perception: given a single RGB image and a natural language instruction, the agent must output pan, tilt, and zoom adjustments of a real PTZ (pan-tilt-zoom) camera to acquire the most informative view for the specified task. We propose EyeVLA, a unified framework that addresses this task by integrating visual perception, language understanding, and physical camera control within a single autoregressive vision-language-action model. EyeVLA introduces a semantically rich and efficient hierarchical action encoding that compactly tokenizes continuous camera adjustments and embeds them into the VLM vocabulary for joint multimodal reasoning. Through a data-efficient pipeline comprising pseudo-label generation, iterative IoU-controlled data refinement, and reinforcement learning with Group Relative Policy Optimization (GRPO), we transfer the open-world understanding of a pre-trained VLM to an embodied active perception policy using only 500 real-world samples. Evaluations on 50 diverse real-world scenes across five independent evaluation runs demonstrate that EyeVLA achieves an average task completion rate of 96%. Our work establishes a new paradigm for instruction-driven active visual information acquisition in multimodal embodied systems. |
|
| ExploreVLA: Dense World Modeling and Exploration for End-to-End Autonomous Driving | 2026-04-03 | ShowEnd-to-end autonomous driving models based on Vision-Language-Action (VLA) architectures have shown promising results by learning driving policies through behavior cloning on expert demonstrations. However, imitation learning inherently limits the model to replicating observed behaviors without exploring diverse driving strategies, leaving it brittle in novel or out-of-distribution scenarios. Reinforcement learning (RL) offers a natural remedy by enabling policy exploration beyond the expert distribution. Yet VLA models, typically trained on offline datasets, lack directly observable state transitions, necessitating a learned world model to anticipate action consequences. In this work, we propose a unified understanding-and-generation framework that leverages world modeling to simultaneously enable meaningful exploration and provide dense supervision. Specifically, we augment trajectory prediction with future RGB and depth image generation as dense world modeling objectives, requiring the model to learn fine-grained visual and geometric representations that substantially enrich the planning backbone. Beyond serving as a supervisory signal, the world model further acts as a source of intrinsic reward for policy exploration: its image prediction uncertainty naturally measures a trajectory's novelty relative to the training distribution, where high uncertainty indicates out-of-distribution scenarios that, if safe, represent valuable learning opportunities. We incorporate this exploration signal into a safety-gated reward and optimize the policy via Group Relative Policy Optimization (GRPO). Experiments on the NAVSIM and nuScenes benchmarks demonstrate the effectiveness of our approach, achieving a state-of-the-art PDMS score of 93.7 and an EPDMS of 88.8 on NAVSIM. The code and demo will be publicly available at https://zihaosheng.github.io/ExploreVLA/. |
The c...The code and demo will be publicly available at https://zihaosheng.github.io/ExploreVLA/ |
| UAV-Track VLA: Embodied Aerial Tracking via Vision-Language-Action Models | 2026-04-02 | ShowEmbodied visual tracking is crucial for Unmanned Aerial Vehicles (UAVs) executing complex real-world tasks. In dynamic urban scenarios with complex semantic requirements, Vision-Language-Action (VLA) models show great promise due to their cross-modal fusion and continuous action generation capabilities. To benchmark multimodal tracking in such environments, we construct a dedicated evaluation benchmark and a large-scale dataset encompassing over 890K frames, 176 tasks, and 85 diverse objects. Furthermore, to address temporal feature redundancy and the lack of spatial geometric priors in existing VLA models, we propose an improved VLA tracking model, UAV-Track VLA. Built upon the |
|
| UniDriveVLA: Unifying Understanding, Perception, and Action Planning for Autonomous Driving | 2026-04-02 | ShowVision-Language-Action (VLA) models have recently emerged in autonomous driving, with the promise of leveraging rich world knowledge to improve the cognitive capabilities of driving systems. However, adapting such models for driving tasks currently faces a critical dilemma between spatial perception and semantic reasoning. Consequently, existing VLA systems are forced into suboptimal compromises: directly adopting 2D Vision-Language Models yields limited spatial perception, whereas enhancing them with 3D spatial representations often impairs the native reasoning capacity of VLMs. We argue that this dilemma largely stems from the coupled optimization of spatial perception and semantic reasoning within shared model parameters. To overcome this, we propose UniDriveVLA, a Unified Driving Vision-Language-Action model based on Mixture-of-Transformers that addresses the perception-reasoning conflict via expert decoupling. Specifically, it comprises three experts for driving understanding, scene perception, and action planning, which are coordinated through masked joint attention. In addition, we combine a sparse perception paradigm with a three-stage progressive training strategy to improve spatial perception while maintaining semantic reasoning capability. Extensive experiments show that UniDriveVLA achieves state-of-the-art performance in open-loop evaluation on nuScenes and closed-loop evaluation on Bench2Drive. Moreover, it demonstrates strong performance across a broad range of perception, prediction, and understanding tasks, including 3D detection, online mapping, motion forecasting, and driving-oriented VQA, highlighting its broad applicability as a unified model for autonomous driving. Code and model have been released at https://github.com/xiaomi-research/unidrivevla |
code ...code has been released at https://github.com/xiaomi-research/unidrivevla |
| DriveDreamer-Policy: A Geometry-Grounded World-Action Model for Unified Generation and Planning | 2026-04-02 | ShowRecently, world-action models (WAM) have emerged to bridge vision-language-action (VLA) models and world models, unifying their reasoning and instruction-following capabilities and spatio-temporal world modeling. However, existing WAM approaches often focus on modeling 2D appearance or latent representations, with limited geometric grounding-an essential element for embodied systems operating in the physical world. We present DriveDreamer-Policy, a unified driving world-action model that integrates depth generation, future video generation, and motion planning within a single modular architecture. The model employs a large language model to process language instructions, multi-view images, and actions, followed by three lightweight generators that produce depth, future video, and actions. By learning a geometry-aware world representation and using it to guide both future prediction and planning within a unified framework, the proposed model produces more coherent imagined futures and more informed driving actions, while maintaining modularity and controllable latency. Experiments on the Navsim v1 and v2 benchmarks demonstrate that DriveDreamer-Policy achieves strong performance on both closed-loop planning and world generation tasks. In particular, our model reaches 89.2 PDMS on Navsim v1 and 88.7 EPDMS on Navsim v2, outperforming existing world-model-based approaches while producing higher-quality future video and depth predictions. Ablation studies further show that explicit depth learning provides complementary benefits to video imagination and improves planning robustness. |
11 pa...11 pages, 4 figures; Project Website: https://drivedreamer-policy.github.io/ |
| Causal Scene Narration with Runtime Safety Supervision for Vision-Language-Action Driving | 2026-04-02 | ShowVision-Language-Action (VLA) models for autonomous driving must integrate diverse textual inputs, including navigation commands, hazard warnings, and traffic state descriptions, yet current systems often present these as disconnected fragments, forcing the model to discover on its own which environmental constraints are relevant to the current maneuver. We introduce Causal Scene Narration (CSN), which restructures VLA text inputs through intent-constraint alignment, quantitative grounding, and structured separation, at inference time with zero GPU cost. We complement CSN with Simplex-based runtime safety supervision and training-time alignment via Plackett-Luce DPO with negative log-likelihood (NLL) regularization. A multi-town closed-loop CARLA evaluation shows that CSN improves Driving Score by +31.1% on original LMDrive and +24.5% on the preference-aligned variant. A controlled ablation reveals that causal structure accounts for 39.1% of this gain, with the remainder attributable to information content alone. A perception noise ablation confirms that CSN's benefit is robust to realistic sensing errors. Semantic safety supervision improves Infraction Score, while reactive Time-To-Collision monitoring degrades performance, demonstrating that intent-aware monitoring is needed for VLA systems. |
18 pa...18 pages, 6 figures, 4 tables |
| Tex3D: Objects as Attack Surfaces via Adversarial 3D Textures for Vision-Language-Action Models | 2026-04-02 | ShowVision-language-action (VLA) models have shown strong performance in robotic manipulation, yet their robustness to physically realizable adversarial attacks remains underexplored. Existing studies reveal vulnerabilities through language perturbations and 2D visual attacks, but these attack surfaces are either less representative of real deployment or limited in physical realism. In contrast, adversarial 3D textures pose a more physically plausible and damaging threat, as they are naturally attached to manipulated objects and are easier to deploy in physical environments. Bringing adversarial 3D textures to VLA systems is nevertheless nontrivial. A central obstacle is that standard 3D simulators do not provide a differentiable optimization path from the VLA objective function back to object appearance, making it difficult to optimize through an end-to-end manner. To address this, we introduce Foreground-Background Decoupling (FBD), which enables differentiable texture optimization through dual-renderer alignment while preserving the original simulation environment. To further ensure that the attack remains effective across long-horizon and diverse viewpoints in the physical world, we propose Trajectory-Aware Adversarial Optimization (TAAO), which prioritizes behaviorally critical frames and stabilizes optimization with a vertex-based parameterization. Built on these designs, we present Tex3D, the first framework for end-to-end optimization of 3D adversarial textures directly within the VLA simulation environment. Experiments in both simulation and real-robot settings show that Tex3D significantly degrades VLA performance across multiple manipulation tasks, achieving task failure rates of up to 96.7%. Our empirical results expose critical vulnerabilities of VLA systems to physically grounded 3D adversarial attacks and highlight the need for robustness-aware training. |
|
| Boosting Vision-Language-Action Finetuning with Feasible Action Neighborhood Prior | 2026-04-02 | ShowIn real-world robotic manipulation, states typically admit a neighborhood of near-equivalent actions. That is, for each state, there exist a feasible action neighborhood (FAN) rather than a single correct action, within which motions yield indistinguishable progress. However, prevalent VLA training methodologies are directly inherited from linguistic settings and do not exploit the FAN property, thus leading to poor generalization and low sample efficiency. To address this limitation, we introduce a FAN-guided regularizer that shapes the model's output distribution to align with the geometry of FAN. Concretely, we introduce a Gaussian prior that promotes locally smooth and unimodal predictions around the preferred direction and magnitude. In extensive experiments across both reinforced finetuning (RFT) and supervised finetuning (SFT), our method achieves significant improvement in sample efficiency, and success rate in both in-distribution and out-of-distribution (OOD) scenarios. By aligning with the intrinsic action tolerance of physical manipulation, FAN-guided regularization provides a principled and practical method for sample-efficient, and generalizable VLA adaptation. |
Accep...Accepted by CVPR 2026 |
| AffordTissue: Dense Affordance Prediction for Tool-Action Specific Tissue Interaction | 2026-04-01 | ShowSurgical action automation has progressed rapidly toward achieving surgeon-like dexterous control, driven primarily by advances in learning from demonstration and vision-language-action models. While these have demonstrated success in table-top experiments, translating them to clinical deployment remains challenging: current methods offer limited predictability on where instruments will interact on tissue surfaces and lack explicit conditioning inputs to enforce tool-action-specific safe interaction regions. Addressing this gap, we introduce AffordTissue, a multimodal framework for predicting tool-action specific tissue affordance regions as dense heatmaps during cholecystectomy. Our approach combines a temporal vision encoder capturing tool motion and tissue dynamics across multiple viewpoints, language conditioning enabling generalization across diverse instrument-action pairs, and a DiT-style decoder for dense affordance prediction. We establish the first tissue affordance benchmark by curating and annotating 15,638 video clips across 103 cholecystectomy procedures, covering six unique tool-action pairs involving four instruments (hook, grasper, scissors, clipper) and their associated tasks: dissection, grasping, clipping, and cutting. Experiments demonstrate substantial improvement over vision-language model baselines (20.6 px ASSD vs. 60.2 px for Molmo-VLM), showing that our task-specific architecture outperforms large-scale foundation models for dense surgical affordance prediction. By predicting tool-action specific tissue affordance regions, AffordTissue provides explicit spatial reasoning for safe surgical automation, potentially unlocking explicit policy guidance toward appropriate tissue regions and early safe stop when instruments deviate outside predicted safe zones. |
|
| RoboNeuron: A Middle-Layer Infrastructure for Agent-Driven Orchestration in Embodied AI | 2026-04-01 | ShowVision-language-action (VLA) models and LLM agents have advanced rapidly, yet reliable deployment on physical robots is often hindered by an interface mismatch between agent tool APIs and robot middleware. Current implementations typically rely on ad-hoc wrappers that are difficult to reuse, and changes to the VLA backend or serving stack often necessitate extensive re-integration. We introduce RoboNeuron, a middleware layer that connects the Model Context Protocol (MCP) for LLM agents with robot middleware such as ROS2. RoboNeuron bridges these ecosystems by deriving agent-callable tools directly from ROS schemas, providing a unified execution abstraction that supports both direct commands and modular composition, and localizing backend, runtime, and acceleration-preset changes within a stable inference boundary. We evaluate RoboNeuron in simulation and on hardware through multi-platform base control, arm motion, and VLA-based grasping tasks, demonstrating that it enables modular system orchestration under a unified interface while supporting backend transitions without system rewiring. The full code implementation of this work is available at github repo: https://github.com/guanweifan/RoboNeuron |
|
| RoboClaw: An Agentic Framework for Scalable Long-Horizon Robotic Tasks | 2026-04-01 | ShowVision-Language-Action (VLA) systems have shown strong potential for language-driven robotic manipulation. However, scaling them to long-horizon tasks remains challenging. Existing pipelines typically separate data collection, policy learning, and deployment, resulting in heavy reliance on manual environment resets and brittle multi-policy execution. We present RoboClaw, an agentic robotics framework that unifies data collection, policy learning, and task execution under a single VLM-driven controller. At the policy level, RoboClaw introduces Entangled Action Pairs (EAP), which couple forward manipulation behaviors with inverse recovery actions to form self-resetting loops for autonomous data collection. This mechanism enables continuous on-policy data acquisition and iterative policy refinement with minimal human intervention. During deployment, the same agent performs high-level reasoning and dynamically orchestrates learned policy primitives to accomplish long-horizon tasks. By maintaining consistent contextual semantics across collection and execution, RoboClaw reduces mismatch between the two phases and improves multi-policy robustness. Experiments in real-world manipulation tasks demonstrate improved stability and scalability compared to conventional open-loop pipelines, while significantly reducing human effort throughout the robot lifecycle, achieving a 25% improvement in success rate over baseline methods on long-horizon tasks and reducing human time investment by 53.7%. |
Code ...Code available at: https://github.com/RoboClaw-Robotics/RoboClaw |
| DVGT-2: Vision-Geometry-Action Model for Autonomous Driving at Scale | 2026-04-01 | ShowEnd-to-end autonomous driving has evolved from the conventional paradigm based on sparse perception into vision-language-action (VLA) models, which focus on learning language descriptions as an auxiliary task to facilitate planning. In this paper, we propose an alternative Vision-Geometry-Action (VGA) paradigm that advocates dense 3D geometry as the critical cue for autonomous driving. As vehicles operate in a 3D world, we think dense 3D geometry provides the most comprehensive information for decision-making. However, most existing geometry reconstruction methods (e.g., DVGT) rely on computationally expensive batch processing of multi-frame inputs and cannot be applied to online planning. To address this, we introduce a streaming Driving Visual Geometry Transformer (DVGT-2), which processes inputs in an online manner and jointly outputs dense geometry and trajectory planning for the current frame. We employ temporal causal attention and cache historical features to support on-the-fly inference. To further enhance efficiency, we propose a sliding-window streaming strategy and use historical caches within a certain interval to avoid repetitive computations. Despite the faster speed, DVGT-2 achieves superior geometry reconstruction performance on various datasets. The same trained DVGT-2 can be directly applied to planning across diverse camera configurations without fine-tuning, including closed-loop NAVSIM and open-loop nuScenes benchmarks. |
Code ...Code is available at \href{https://github.com/wzzheng/DVGT} |
| Do World Action Models Generalize Better than VLAs? A Robustness Study | 2026-04-01 | ShowRobot action planning in the real world is challenging as it requires not only understanding the current state of the environment but also predicting how it will evolve in response to actions. Vision-language-action (VLA), which repurpose large-scale vision-language models for robot action generation using action experts, have achieved notable success across a variety of robotic tasks. Nevertheless, their performance remains constrained by the scope of their training data, exhibiting limited generalization to unseen scenarios and vulnerability to diverse contextual perturbations. More recently, world models have been revisited as an alternative to VLAs. These models, referred to as world action models (WAMs), are built upon world models that are trained on large corpora of video data to predict future states. With minor adaptations, their latent representation can be decoded into robot actions. It has been suggested that their explicit dynamic prediction capacity, combined with spatiotemporal priors acquired from web-scale video pretraining, enables WAMs to generalize more effectively than VLAs. In this paper, we conduct a comparative study of prominent state-of-the-art VLA policies and recently released WAMs. We evaluate their performance on the LIBERO-Plus and RoboTwin 2.0-Plus benchmarks under various visual and language perturbations. Our results show that WAMs achieve strong robustness, with LingBot-VA reaching 74.2% success rate on RoboTwin 2.0-Plus and Cosmos-Policy achieving 82.2% on LIBERO-Plus. While VLAs such as |
|
| DIAL: Decoupling Intent and Action via Latent World Modeling for End-to-End VLA | 2026-03-31 | ShowThe development of Vision-Language-Action (VLA) models has been significantly accelerated by pre-trained Vision-Language Models (VLMs). However, most existing end-to-end VLAs treat the VLM primarily as a multimodal encoder, directly mapping vision-language features to low-level actions. This paradigm underutilizes the VLM's potential in high-level decision making and introduces training instability, frequently degrading its rich semantic representations. To address these limitations, we introduce DIAL, a framework bridging high-level decision making and low-level motor execution through a differentiable latent intent bottleneck. Specifically, a VLM-based System-2 performs latent world modeling by synthesizing latent visual foresight within the VLM's native feature space; this foresight explicitly encodes intent and serves as the structural bottleneck. A lightweight System-1 policy then decodes this predicted intent together with the current observation into precise robot actions via latent inverse dynamics. To ensure optimization stability, we employ a two-stage training paradigm: a decoupled warmup phase where System-2 learns to predict latent futures while System-1 learns motor control under ground-truth future guidance within a unified feature space, followed by seamless end-to-end joint optimization. This enables action-aware gradients to refine the VLM backbone in a controlled manner, preserving pre-trained knowledge. Extensive experiments on the RoboCasa GR1 Tabletop benchmark show that DIAL establishes a new state-of-the-art, achieving superior performance with 10x fewer demonstrations than prior methods. Furthermore, by leveraging heterogeneous human demonstrations, DIAL learns physically grounded manipulation priors and exhibits robust zero-shot generalization to unseen objects and novel configurations during real-world deployment on a humanoid robot. |
Proje...Project page: https://xpeng-robotics.github.io/dial |
| DFM-VLA: Iterative Action Refinement for Robot Manipulation via Discrete Flow Matching | 2026-03-31 | ShowVision--Language--Action (VLA) models that encode actions using a discrete tokenization scheme are increasingly adopted for robotic manipulation, but existing decoding paradigms remain fundamentally limited. Whether actions are decoded sequentially by autoregressive VLAs or in parallel by discrete diffusion VLAs, once a token is generated, it is typically fixed and cannot be revised in subsequent iterations, so early token errors cannot be effectively corrected later. We propose DFM-VLA, a discrete flow matching VLA for iterative refinement of action tokens. DFM-VLA~models a token-level probability velocity field that dynamically updates the full action sequence across refinement iterations. We investigate two ways to construct the velocity field: an auxiliary velocity-head formulation and an action-embedding-guided formulation. Our framework further adopts a two-stage decoding strategy with an iterative refinement stage followed by deterministic validation for stable convergence. Extensive experiments on CALVIN, LIBERO, and real-world manipulation tasks show that DFM-VLA consistently outperforms strong autoregressive, discrete diffusion, and continuous diffusion baselines in manipulation performance while retaining high inference efficiency. In particular, DFM-VLA achieves an average success length of 4.44 on CALVIN and an average success rate of 95.7% on LIBERO, highlighting the value of action refinement via discrete flow matching for robotic manipulation. Our project is available https://chris1220313648.github.io/DFM-VLA/ |
|
| CoMaTrack: Competitive Multi-Agent Game-Theoretic Tracking with Vision-Language-Action Models | 2026-03-31 | ShowEmbodied Visual Tracking (EVT), a core dynamic task in embodied intelligence, requires an agent to precisely follow a language-specified target. Yet most existing methods rely on single-agent imitation learning, suffering from costly expert data and limited generalization due to static training environments. Inspired by competition-driven capability evolution, we propose CoMaTrack, a competitive game-theoretic multi-agent reinforcement learning framework that trains agents in a dynamic adversarial setting with competitive subtasks, yielding stronger adaptive planning and interference-resilient strategies. We further introduce CoMaTrack-Bench, the first open-source Habitat-based benchmark protocol and episode set for language-conditioned competitive EVT featuring dynamic dueling, featuring game scenarios between a tracker and adaptive opponents across diverse environments and instructions, enabling standardized robustness evaluation under active adversarial interactions. Experiments show that CoMaTrack achieves state-of-the-art results on both standard benchmarks and CoMaTrack-Bench. Notably, a 3B VLM trained with our framework surpasses previous single-agent imitation learning methods based on 7B models on the challenging EVT-Bench, achieving 92.1% in STT, 74.2% in DT, and 57.5% in AT. The benchmark code will be available at https://github.com/wlqcode/CoMaTrack-Bench. |
|
| X-World: Controllable Ego-Centric Multi-Camera World Models for Scalable End-to-End Driving | 2026-03-31 | ShowScalable and reliable evaluation is increasingly critical in the end-to-end era of autonomous driving, where vision--language--action (VLA) policies directly map raw sensor streams to driving actions. Yet, current evaluation pipelines still rely heavily on real-world road testing, which is costly, biased toward limited scenario coverage, and difficult to reproduce. These challenges motivate a real-world simulator that can generate realistic future observations under proposed actions, while remaining controllable and stable over long horizons. We present X-World, an action-conditioned multi-camera generative world model that simulates future observations directly in video space. Given synchronized multi-view camera history and a future action sequence, X-World generates future multi-camera video streams that follow the commanded actions. To ensure reproducible and editable scene rollouts, X-World further supports optional controls over dynamic traffic agents and static road elements, and retains a text-prompt interface for appearance-level control (e.g., weather and time of day). Beyond world simulation, X-World also enables video style transfer by conditioning on appearance prompts while preserving the underlying action and scene dynamics. At the core of X-World is a multi-view latent video generator designed to explicitly encourage cross-view geometric consistency and temporal coherence under diverse control signals. Experiments show that X-World achieves high-quality multi-view video generation with (i) strong view consistency across cameras, (ii) stable temporal dynamics over long rollouts, and (iii) high controllability with strict action following and faithful adherence to optional scene controls. These properties make X-World a practical foundation for scalable and reproducible evaluation. |
Technical Report |
| VLA Models Are More Generalizable Than You Think: Revisiting Physical and Spatial Modeling | 2026-03-31 | ShowVision-language-action (VLA) models achieve strong in-distribution performance but degrade sharply under novel camera viewpoints and visual perturbations. We show that this brittleness primarily arises from misalignment in Spatial Modeling, rather than Physical Modeling. To address this, we propose a one-shot adaptation framework that recalibrates visual representations through lightweight, learnable updates. Our first method, Feature Token Modulation (FTM), applies a global affine transformation to visual tokens and improves Libero viewpoint accuracy from 48.5% to 87.1% with only 4K parameters. Building on this, Feature Linear Adaptation (FLA) introduces low-rank updates to the ViT encoder, achieving 90.8% success with 4.7M parameters -- matching LoRA-scale finetuning at far lower cost. Together, these results reveal substantial untapped robustness in pretrained VLA models and demonstrate that targeted, minimal visual adaptation is sufficient to restore viewpoint generalization. |
|
| FocusVLA: Focused Visual Utilization for Vision-Language-Action Models | 2026-03-30 | ShowVision-Language-Action (VLA) models improve action generation by conditioning policies on rich vision-language information. However, current auto-regressive policies are constrained by three bottlenecks: (1) architectural bias drives models to overlook visual details, (2) an excessive number of visual tokens makes attention difficult to focus on the correct regions, and (3) task-irrelevant visual information introduces substantial noise - together severely impairing the quality of action. In this paper, we investigate how to effectively utilize different visual representations for action generation. To this end, we first empirically validate the above issues and show that VLA performance is primarily limited by how visual information is utilized, rather than by the quality of visual representations. Based on these insights, we introduce FocusVLA, a novel paradigm that directs the model's attention to task-relevant visual regions to effectively bridge vision to action. Specifically, we first propose Modality Cascaded Attention to eliminate shortcut pathways, thereby compelling VLA models to rely on task-relevant visual details for action generation. Furthermore, we propose Focus Attention, which dynamically selects task-relevant visual patches to control information quantity while explicitly modulating their influence to suppress task-irrelevant noise. Extensive experiments on both simulated and real-world robotic benchmarks demonstrate that FocusVLA not only effectively leverages visual details to perform dexterous manipulations, but also substantially improves performance and accelerates convergence across a variety of tasks. |
25 pages, 18 figures |
| ACoT-VLA: Action Chain-of-Thought for Vision-Language-Action Models | 2026-03-30 | ShowVision-Language-Action models have emerged as essential generalist robot policies for diverse manipulation tasks, conventionally relying on directly translating multimodal inputs into actions via Vision-Language Model embeddings. Recent advancements have introduced explicit intermediary reasoning-such as sub-task prediction (language) or goal image synthesis (vision)-to guide action generation. However, these intermediate reasoning are often indirect and inherently limited in their capacity to convey the full, granular information required for precise action execution. Instead, we posit that the most effective form of reasoning is one that deliberates directly in the action space. We introduce Action Chain-of-Thought (ACoT), a paradigm where the reasoning process itself is formulated as a structured sequence of coarse action intents that guide the final policy. In this paper, we propose ACoT-VLA, a novel architecture that materializes the ACoT paradigm. Specifically, we introduce two complementary components: an Explicit Action Reasoner (EAR) and Implicit Action Reasoner (IAR). The former proposes coarse reference trajectories as explicit action-level reasoning steps, while the latter extracts latent action priors from internal representations of multimodal input, co-forming an ACoT that conditions the downstream action head to enable grounded policy learning. Extensive experiments in real-world and simulation environments demonstrate the superiority of our proposed method. Code is available at: https://github.com/AgibotTech/ACoT-VLA. |
Accep...Accepted by Conference on Computer Vision and Pattern Recognition (CVPR) 2026 |
| 3D CAVLA: Leveraging Depth and 3D Context to Generalize Vision Language Action Models for Unseen Tasks | 2026-03-30 | ShowRobotic manipulation in 3D requires effective computation of N degree-of-freedom joint-space trajectories that enable precise and robust control. To achieve this, robots must integrate semantic understanding with visual perception to transform real-world observations into low-level control for object interaction. Recent advances in Vision-Language-Action (VLA) models have shown promise by mapping RGB images and language instructions to task space velocities, typically trained on large datasets of teleoperated demonstrations. However, these models often struggle with generalization beyond their training distributions. In this work, we introduce 3D-CAVLA, a novel finetuning framework that enhances task generalization of VLA policies by incorporating three key components: (i) chain-of-thought reasoning for structured decision-making, (ii) depth-aware perception for 3D spatial understanding, and (iii) task-oriented region-of-interest detection for focused manipulation. Extensive experiments in the LIBERO simulation environment demonstrate that 3D-CAVLA achieves an average success rate of 98.1% across diverse in-domain task suites. On unseen tasks, 3D-CAVLA delivers an absolute improvement of 8.8% in success rate, underscoring the benefits of 3D scene awareness for robust generalization. We validate our approach on real-world tabletop experiments demonstrating that the proposed model translates effectively from simulation to physical robots. 3D-CAVLA achieves over a 3X faster training convergence and delivers a 25% gain in success rate on unseen real world tasks. We will open-source our code and the unseen tasks dataset to promote community-driven research here: https://3d-cavla.github.io |
Accep...Accepted at the 1st Workshop on 3D LLM/VLA, CVPR 2025. This work has been submitted to the IEEE for possible publication |
| StreamingVLA: Streaming Vision-Language-Action Model with Action Flow Matching and Adaptive Early Observation | 2026-03-30 | ShowVision-language-action (VLA) models have demonstrated exceptional performance in natural language-driven perception and control. However, the high computational cost of VLA models poses significant efficiency challenges, particularly for resource-constrained edge platforms in real-world deployments. However, since different stages of VLA (observation, action generation and execution) must proceed sequentially, and wait for the completion of the preceding stage, the system suffers from frequent halting and high latency. To address this, We conduct a systematic analysis to identify the challenges for fast and fluent generation, and propose enabling VLAs with the ability to asynchronously parallelize across VLA stages in a "streaming" manner. First, we eliminate the reliance on action chunking and adopt action flow matching, which learns the trajectory of action flows rather than denoising chunk-wise actions. It overlaps the latency of action generation and execution. Second, we design an action saliency-aware adaptive observation mechanism, thereby overlapping the latency of execution and observation. Without sacrificing performance, StreamingVLA achieves substantial speedup and improves the fluency of execution. It achieves a 2.4 |
|
| ManipArena: Comprehensive Real-world Evaluation of Reasoning-Oriented Generalist Robot Manipulation | 2026-03-30 | ShowVision-Language-Action (VLA) models and world models have recently emerged as promising paradigms for general-purpose robotic intelligence, yet their progress is hindered by the lack of reliable evaluation protocols that reflect real-world deployment. Existing benchmarks are largely simulator-centric, which provide controllability but fail to capture the reality gap caused by perception noise, complex contact dynamics, hardware constraints, and system latency. Moreover, fragmented real-world evaluations across different robot platforms prevent fair and reproducible comparison. To address these challenges, we introduce ManipArena, a standardized evaluation framework designed to bridge simulation and real-world execution. ManipArena comprises 20 diverse tasks across 10,812 expert trajectories emphasizing reasoning-oriented manipulation tasks requiring semantic and spatial reasoning, supports multi-level generalization through controlled out-of-distribution settings, and incorporates long-horizon mobile manipulation beyond tabletop scenarios. The framework further provides rich sensory diagnostics, including low-level motor signals, and synchronized real-to-sim environments constructed via high-quality 3D scanning. Together, these features enable fair, realistic, and reproducible evaluation for both VLA and world model approaches, providing a scalable foundation for diagnosing and advancing embodied intelligence systems. |
Techn...Technical report for CVPR 2026 Challenge ManipArena |
| From Observation to Action: Latent Action-based Primitive Segmentation for VLA Pre-training in Industrial Settings | 2026-03-30 | ShowWe present a novel unsupervised framework to unlock vast unlabeled human demonstration data from continuous industrial video streams for Vision-Language-Action (VLA) model pre-training. Our method first trains a lightweight motion tokenizer to encode motion dynamics, then employs an unsupervised action segmenter leveraging a novel "Latent Action Energy" metric to discover and segment semantically coherent action primitives. The pipeline outputs both segmented video clips and their corresponding latent action sequences, providing structured data directly suitable for VLA pre-training. Evaluations on public benchmarks and a proprietary electric motor assembly dataset demonstrate effective segmentation of key tasks performed by humans at workstations. Further clustering and quantitative assessment via a Vision-Language Model confirm the semantic coherence of the discovered action primitives. To our knowledge, this is the first fully automated end-to-end system for extracting and organizing VLA pre-training data from unstructured industrial videos, offering a scalable solution for embodied AI integration in manufacturing. |
10 pa...10 pages, 5 figures, Accepted to CVPR 2026 |
| LIBERO-Para: A Diagnostic Benchmark and Metrics for Paraphrase Robustness in VLA Models | 2026-03-30 | ShowVision-Language-Action (VLA) models achieve strong performance in robotic manipulation by leveraging pre-trained vision-language backbones. However, in downstream robotic settings, they are typically fine-tuned with limited data, leading to overfitting to specific instruction formulations and leaving robustness to paraphrased instructions underexplored. To study this gap, we introduce LIBERO-Para, a controlled benchmark that independently varies action expressions and object references for fine-grained analysis of linguistic generalization. Across seven VLA configurations (0.6B-7.5B), we observe consistent performance degradation of 22-52 pp under paraphrasing. This degradation is primarily driven by object-level lexical variation: even simple synonym substitutions cause large drops, indicating reliance on surface-level matching rather than semantic grounding. Moreover, 80-96% of failures arise from planning-level trajectory divergence rather than execution errors, showing that paraphrasing disrupts task identification. Binary success rate treats all paraphrases equally, obscuring whether models perform consistently across difficulty levels or rely on easier cases. To address this, we propose PRIDE, a metric that quantifies paraphrase difficulty using semantic and syntactic factors. Our benchmark and corresponding code are available at: https://github.com/cau-hai-lab/LIBERO-Para |
32 pages, 28 figures |
| Vega: Learning to Drive with Natural Language Instructions | 2026-03-30 | ShowVision-language-action models have reshaped autonomous driving to incorporate languages into the decision-making process. However, most existing pipelines only utilize the language modality for scene descriptions or reasoning and lack the flexibility to follow diverse user instructions for personalized driving. To address this, we first construct a large-scale driving dataset (InstructScene) containing around 100,000 scenes annotated with diverse driving instructions with the corresponding trajectories. We then propose a unified Vision-Language-World-Action model, Vega, for instruction-based generation and planning. We employ the autoregressive paradigm to process visual inputs (vision) and language instructions (language) and the diffusion paradigm to generate future predictions (world modeling) and trajectories (action). We perform joint attention to enable interactions between the modalities and use individual projection layers for different modalities for more capabilities. Extensive experiments demonstrate that our method not only achieves superior planning performance but also exhibits strong instruction-following abilities, paving the way for more intelligent and personalized driving systems. |
Code ...Code is available at https://github.com/zuosc19/Vega |
| ThermoAct:Thermal-Aware Vision-Language-Action Models for Robotic Perception and Decision-Making | 2026-03-30 | ShowIn recent human-robot collaboration environments, there is a growing focus on integrating diverse sensor data beyond visual information to enable safer and more intelligent task execution. Although thermal data can be crucial for enhancing robot safety and operational efficiency, its integration has been relatively overlooked in prior research. This paper proposes a novel Vision-Language-Action (VLA) framework that incorporates thermal information for robot task execution. The proposed system leverages a Vision-Language Model (VLM) as a high-level planner to interpret complex natural language commands and decompose them into simpler sub-tasks. This approach facilitates efficient data collection and robust reasoning for complex operations. Unlike conventional methods that rely solely on visual data, our approach integrates thermal information, enabling the robot to perceive physical properties and proactively ensure environmental safety. Experimental results from real-world task scenarios validate the feasibility of our proposed framework, suggesting its potential to enhance task success rates and safety compared to existing vision-based systems. |
2026 RA-L |
| LaST$_{0}$: Latent Spatio-Temporal Chain-of-Thought for Robotic Vision-Language-Action Model | 2026-03-30 | ShowVision-Language-Action (VLA) models have recently shown strong generalization, with some approaches seeking to explicitly generate linguistic reasoning traces or predict future observations prior to execution. However, explicit reasoning typically incurs non-negligible inference latency, which constrains the temporal resolution required for robotic manipulation. Moreover, such reasoning is confined to the linguistic space, imposing a representational bottleneck that struggles to faithfully capture ineffable physical attributes. To mitigate these limitations, we propose LaST$_0$, a framework that enables efficient reasoning before acting through a Latent Spatio-Temporal Chain-of-Thought (CoT), capturing fine-grained physical and robotic dynamics that are often difficult to verbalize. Specifically, we introduce a token-efficient latent CoT space that models future visual dynamics, 3D structural information, and robot proprioceptive states, and further extends these representations across time to enable temporally consistent implicit reasoning trajectories. Furthermore, LaST$_0$ adopts a dual-system architecture implemented via a Mixture-of-Transformers design, where a reasoning expert conducts low-frequency latent inference and an acting expert generates high-frequency actions conditioned on robotics-oriented latent representations. To facilitate coordination, LaST$_0$ is trained with heterogeneous operation frequencies, enabling adaptive switching during deployment. Across 10 real-world tasks spanning tabletop, mobile, and dexterous hand manipulation, LaST$_0$ improves mean success rates by 13%, 14% and 14% over prior SOTA VLA methods, respectively. |
Proje...Project page: https://vla-last0.github.io/ |
| ROBOGATE: Adaptive Failure Discovery for Safe Robot Policy Deployment via Two-Stage Boundary-Focused Sampling | 2026-03-30 | ShowDeploying learned robot manipulation policies in industrial settings requires rigorous pre-deployment validation, yet exhaustive testing across high-dimensional parameter spaces is intractable. We present ROBOGATE, a deployment risk management framework that combines physics-based simulation with a two-stage adaptive sampling strategy to efficiently discover failure boundaries in the operational parameter space. Stage 1 employs Latin Hypercube Sampling (LHS) across an 8-dimensional parameter space to establish a coarse failure landscape from 20,000 uniformly distributed experiments. Stage 2 applies boundary-focused sampling that concentrates 10,000 additional experiments in the 30-70% success rate transition zone, enabling precise failure boundary mapping. Using NVIDIA Isaac Sim with Newton physics, we evaluate a scripted pick-and-place controller on two robot embodiments -- Franka Panda (7-DOF) and UR5e (6-DOF) -- across 30,000 total experiments. Our logistic regression risk model achieves an AUC of 0.780 on the combined dataset (vs. 0.754 for Stage 1 alone), identifies a closed-form failure boundary equation, and reveals four universal danger zones affecting both robot platforms. We further demonstrate the framework on VLA (Vision-Language-Action) model evaluation, where Octo-Small achieves 0.0% success rate on 68 adversarial scenarios versus 100% for the scripted baseline -- a 100-point gap that underscores the challenge of deploying foundation models in industrial settings. ROBOGATE is open-source and runs on a single GPU workstation. |
12 pa...12 pages, 5 figures, open-source code and 30K failure pattern dataset available at https://github.com/liveplex-cpu/robogate |
| CARLA-Air: Fly Drones Inside a CARLA World -- A Unified Infrastructure for Air-Ground Embodied Intelligence | 2026-03-30 | ShowThe convergence of low-altitude economies, embodied intelligence, and air-ground cooperative systems creates growing demand for simulation infrastructure capable of jointly modeling aerial and ground agents within a single physically coherent environment. Existing open-source platforms remain domain-segregated: driving simulators lack aerial dynamics, while multirotor simulators lack realistic ground scenes. Bridge-based co-simulation introduces synchronization overhead and cannot guarantee strict spatial-temporal consistency. We present CARLA-Air, an open-source infrastructure that unifies high-fidelity urban driving and physics-accurate multirotor flight within a single Unreal Engine process. The platform preserves both CARLA and AirSim native Python APIs and ROS 2 interfaces, enabling zero-modification code reuse. Within a shared physics tick and rendering pipeline, CARLA-Air delivers photorealistic environments with rule-compliant traffic, socially-aware pedestrians, and aerodynamically consistent UAV dynamics, synchronously capturing up to 18 sensor modalities across all platforms at each tick. The platform supports representative air-ground embodied intelligence workloads spanning cooperation, embodied navigation and vision-language action, multi-modal perception and dataset construction, and reinforcement-learning-based policy training. An extensible asset pipeline allows integration of custom robot platforms into the shared world. By inheriting AirSim's aerial capabilities -- whose upstream development has been archived -- CARLA-Air ensures this widely adopted flight stack continues to evolve within a modern infrastructure. Released with prebuilt binaries and full source: https://github.com/louiszengCN/CarlaAir |
Prebu...Prebuilt binaries, project page, full source code, and community discussion group are all available at: https://github.com/louiszengCN/CarlaAir |
| Learning Multi-View Spatial Reasoning from Cross-View Relations | 2026-03-30 | ShowVision-language models (VLMs) have achieved impressive results on single-view vision tasks, but lack the multi-view spatial reasoning capabilities essential for embodied AI systems to understand 3D environments and manipulate objects across different viewpoints. In this work, we introduce Cross-View Relations (XVR), a large-scale dataset designed to teach VLMs spatial reasoning across multiple views. XVR comprises 100K vision-question-answer samples derived from 18K diverse 3D scenes and 70K robotic manipulation trajectories, spanning three fundamental spatial reasoning tasks: Correspondence (matching objects across views), Verification (validating spatial relationships), and Localization (identifying object positions). VLMs fine-tuned on XVR achieve substantial improvements on established multi-view and robotic spatial reasoning benchmarks (MindCube and RoboSpatial). When integrated as backbones in Vision-Language-Action models, XVR-trained representations improve success rates on RoboCasa. Our results demonstrate that explicit training on cross-view spatial relations significantly enhances multi-view reasoning and transfers effectively to real-world robotic manipulation. |
Accep...Accepted to CVPR 2026 |
| Goal-VLA: Image-Generative VLMs as Object-Centric World Models Empowering Zero-shot Robot Manipulation | 2026-03-30 | ShowGeneralization remains a fundamental challenge in robotic manipulation. To tackle this challenge, recent Vision-Language-Action (VLA) models build policies on top of Vision-Language Models (VLMs), seeking to transfer their open-world semantic knowledge. However, their zero-shot capability lags significantly behind the base VLMs, as the instruction-vision-action data is too limited to cover diverse scenarios, tasks, and robot embodiments. In this work, we present Goal-VLA, a zero-shot framework that leverages Image-Generative VLMs as world models to generate desired goal states, from which the target object pose is derived to enable generalizable manipulation. The key insight is that object state representation is the golden interface, naturally separating a manipulation system into high-level and low-level policies. This representation abstracts away explicit action annotations, allowing the use of highly generalizable VLMs while simultaneously providing spatial cues for training-free low-level control. To further improve robustness, we introduce a Reflection-through-Synthesis process that iteratively validates and refines the generated goal image before execution. Both simulated and real-world experiments demonstrate that our \name achieves strong performance and inspiring generalizability in manipulation tasks. Supplementary materials are available at https://nus-lins-lab.github.io/goalvlaweb/. |
|
| ProgressVLA: Progress-Guided Diffusion Policy for Vision-Language Robotic Manipulation | 2026-03-29 | ShowMost existing vision-language-action (VLA) models for robotic manipulation lack progress awareness, typically relying on hand-crafted heuristics for task termination. This limitation is particularly severe in long-horizon tasks involving cascaded sub-goals. In this work, we investigate the estimation and integration of task progress, proposing a novel model named {\textbf \vla}. Our technical contributions are twofold: (1) \emph{robust progress estimation}: We pre-train a progress estimator on large-scale, unsupervised video-text robotic datasets. This estimator achieves a low prediction residual (0.07 on a scale of |
|
| Uni-World VLA: Interleaved World Modeling and Planning for Autonomous Driving | 2026-03-28 | ShowAutonomous driving requires reasoning about how the environment evolves and planning actions accordingly. Existing world-model-based approaches typically predict future scenes first and plan afterwards, resulting in open-loop imagination that may drift from the actual decision process. In this paper, we present Uni-World VLA, a unified vision-language-action (VLA) model that tightly interleaves future frame prediction and trajectory planning. Instead of generating a full world rollout before planning, our model alternates between predicting future frames and ego actions step by step, allowing planning decisions to be continuously conditioned on the imagined future observations. This interleaved generation forms a closed-loop interaction between world modeling and control, enabling more adaptive decision-making in dynamic traffic scenarios. In addition, we incorporate monocular depth information into frames to provide stronger geometric cues for world modeling, improving long-horizon scene prediction. Experiments on the NAVSIM benchmark show that our approach achieves competitive closed-loop planning performance while producing high-fidelity future frame predictions. These results demonstrate that tightly coupling world prediction and planning is a promising direction for scalable VLA driving systems. |
22 pa...22 pages, 8 figures. Submitted to ECCV 2026. Code will be released |
| CycleManip: Enabling Cyclic Task Manipulation via Effective Historical Perception and Understanding | 2026-03-28 | ShowIn this paper, we explore an important yet underexplored task in robot manipulation: cycle-based manipulation, where robots need to perform cyclic or repetitive actions with an expected terminal time. These tasks are crucial in daily life, such as shaking a bottle or knocking a nail. However, few prior works have explored this task, leading to two main challenges: 1) the imitation methods often fail to complete these tasks within the expected terminal time due to the ineffective utilization of history; 2) the absence of a benchmark with sufficient data and automatic evaluation tools hinders development of effective solutions in this area. To address these challenges, we first propose the CycleManip framework to achieve cycle-based task manipulation in an end-to-end imitation manner without requiring any extra models, hierarchical structure or significant computational overhead. The core insight is to enhance effective history perception by a cost-aware sampling strategy and to improve historical understanding by multi-task learning. Second, we introduce a cycle-based task manipulation benchmark, which provides diverse cycle-based tasks, and an automatic evaluation method. Extensive experiments conducted in both simulation and real-world settings demonstrate that our method achieves high success rates in cycle-based task manipulation. The results further show strong adaptability performance in general manipulation, and the plug-and-play ability on imitation policies such as Vision-Language-Action (VLA) models. Moreover, the results show that our approach can be applied across diverse robotic platforms, including bi-arm grippers, dexterous hands, and humanoid robots. |
Accep...Accepted by CVPR2026. Project page: https://isee-laboratory.github.io/CycleManip/ |
| Scaling Sim-to-Real Reinforcement Learning for Robot VLAs with Generative 3D Worlds | 2026-03-28 | ShowThe strong performance of large vision-language models (VLMs) trained with reinforcement learning (RL) has motivated similar approaches for fine-tuning vision-language-action (VLA) models in robotics. Many recent works fine-tune VLAs directly in the real world to avoid addressing the sim-to-real gap. While real-world RL circumvents sim-to-real issues, it inherently limits the generality of the resulting VLA, as scaling scene and object diversity in the physical world is prohibitively difficult. This leads to the paradoxical outcome of transforming a broadly pretrained model into an overfitted, scene-specific policy. Training in simulation can instead provide access to diverse scenes, but designing those scenes is also costly. In this work, we show that VLAs can be RL fine-tuned without sacrificing generality and with reduced labor by leveraging 3D world generative models. Using these models together with a language-driven scene designer, we generate hundreds of diverse interactive scenes containing unique objects and backgrounds, enabling scalable and highly parallel policy learning. Starting from a pretrained imitation baseline, our approach increases simulation success from 9.7% to 79.8% while achieving a 1.25$\times$ speedup in task completion time. We further demonstrate successful sim-to-real transfer enabled by the quality of the generated digital twins together with domain randomization, improving real-world success from 21.7% to 75% and achieving a 1.13$\times$ speedup. Finally, we further highlight the benefits of leveraging the effectively unlimited data from 3D world generative models through an ablation study showing that increasing scene diversity directly improves zero-shot generalization. |
|
| Mimic Intent, Not Just Trajectories | 2026-03-28 | ShowWhile imitation learning (IL) has achieved impressive success in dexterous manipulation through generative modeling and pretraining, state-of-the-art approaches like Vision-Language-Action (VLA) models still struggle with adaptation to environmental changes and skill transfer. We argue this stems from mimicking raw trajectories without understanding the underlying intent. To address this, we propose explicitly disentangling behavior intent from execution details in end-2-end IL: Mimic Intent, Not just Trajectories(MINT). We achieve this via multi-scale frequency-space tokenization, which enforces a spectral decomposition of action chunk representation. We learn action tokens with a multi-scale coarse-to-fine structure, and force the coarsest token to capture low-frequency global structure and finer tokens to encode high-frequency details. This yields an abstract Intent token that facilitates planning and transfer, and multi-scale Execution tokens that enable precise adaptation to environmental dynamics. Building on this hierarchy, our policy generates trajectories through next-scale autoregression, performing progressive intent-to-execution reasoning, thus boosting learning efficiency and generalization. Crucially, this disentanglement enables one-shot transfer of skills, by simply injecting the Intent token from a demonstration into the autoregressive generation process. Experiments on several manipulation benchmarks and on a real robot demonstrate state-of-the-art success rates, superior inference efficiency, robust generalization against disturbances, and effective one-shot transfer. |
|
| VLA-OPD: Bridging Offline SFT and Online RL for Vision-Language-Action Models via On-Policy Distillation | 2026-03-27 | ShowAlthough pre-trained Vision-Language-Action (VLA) models exhibit impressive generalization in robotic manipulation, post-training remains crucial to ensure reliable performance during deployment. However, standard offline Supervised Fine-Tuning (SFT) suffers from distribution shifts and catastrophic forgetting of pre-trained capabilities, while online Reinforcement Learning (RL) struggles with sparse rewards and poor sample efficiency. In this paper, we propose On-Policy VLA Distillation (VLA-OPD), a framework bridging the efficiency of SFT with the robustness of RL. Instead of relying on sparse environmental rewards, VLA-OPD leverages an expert teacher to provide dense, token-level supervision on the student's self-generated trajectories. This enables active error correction on policy-induced states while preserving pre-trained general capabilities through gentle alignment. Crucially, we formulate VLA-OPD via a Reverse-KL objective. Unlike standard Forward-KL that induces mode-covering entropy explosion, or Hard-CE that causes premature entropy collapse, our bounded mode-seeking objective ensures stable policy learning by filtering out the teacher's epistemic uncertainty while maintaining action diversity. Experiments on LIBERO and RoboTwin2.0 benchmarks demonstrate that VLA-OPD significantly improves sample efficiency over RL and robustness over SFT, while effectively mitigating catastrophic forgetting during post-training. |
|
| MMaDA-VLA: Large Diffusion Vision-Language-Action Model with Unified Multi-Modal Instruction and Generation | 2026-03-27 | ShowVision-Language-Action (VLA) models aim to control robots for manipulation from visual observations and natural-language instructions. However, existing hierarchical and autoregressive paradigms often introduce architectural overhead, suffer from temporal inconsistency and long-horizon error accumulation, and lack a mechanism to capture environment dynamics without extra modules. To this end, we present MMaDA-VLA, a fully native pre-trained large diffusion VLA model that unifies multi-modal understanding and generation in a single framework. Our key idea is a native discrete diffusion formulation that embeds language, images, and continuous robot controls into one discrete token space and trains a single backbone with masked token denoising to jointly generate a future goal observation and an action chunk in parallel. Iterative denoising enables global, order-free refinement, improving long-horizon consistency while grounding actions in predicted future visual outcomes without auxiliary world models. Experiments across simulation benchmarks and real-world tasks show state-of-the-art performance, achieving 98.0% average success on LIBERO and 4.78 average length on CALVIN. |
|
| DiffusionAnything: End-to-End In-context Diffusion Learning for Unified Navigation and Pre-Grasp Motion | 2026-03-27 | ShowEfficiently predicting motion plans directly from vision remains a fundamental challenge in robotics, where planning typically requires explicit goal specification and task-specific design. Recent vision-language-action (VLA) models infer actions directly from visual input but demand massive computational resources, extensive training data, and fail zero-shot in novel scenes. We present a unified image-space diffusion policy handling both meter-scale navigation and centimeter-scale manipulation via multi-scale feature modulation, with only 5 minutes of self-supervised data per task. Three key innovations drive the framework: (1) Multi-scale FiLM conditioning on task mode, depth scale, and spatial attention enables task-appropriate behavior in a single model; (2) trajectory-aligned depth prediction focuses metric 3D reasoning along generated waypoints; (3) self-supervised attention from AnyTraverse enables goal-directed inference without vision-language models and depth sensors. Operating purely from RGB input (2.0 GB memory, 10 Hz), the model achieves robust zero-shot generalization to novel scenes while remaining suitable for onboard deployment. |
|
| SOMA: Strategic Orchestration and Memory-Augmented System for Vision-Language-Action Model Robustness via In-Context Adaptation | 2026-03-27 | ShowDespite the promise of Vision-Language-Action (VLA) models as generalist robotic controllers, their robustness against perceptual noise and environmental variations in out-of-distribution (OOD) tasks remains fundamentally limited by the absence of long-term memory, causal failure attribution, and dynamic intervention capability. To address this, we propose SOMA, a Strategic Orchestration and Memory-Augmented System that upgrades frozen VLA policies for robust in-context adaptation without parameter fine-tuning. Specifically, SOMA operates through an online pipeline of contrastive Dual-Memory Retrieval-Augmented Generation (RAG), an Attribution-Driven Large-Language-Model (LLM) Orchestrator, and extensible Model Context Protocol (MCP) interventions, while an offline Memory Consolidation module continuously distills the execution traces into reliable priors. Experimental evaluations across three backbone models (pi0, pi0.5, and SmolVLA) on LIBERO-PRO and our proposed LIBERO-SOMA benchmarks demonstrate that SOMA achieves an average absolute success rate gain of 56.6%. This includes a significant absolute improvement of 89.1% in long-horizon task chaining. Project page and source code are available at: https://github.com/LZY-1021/SOMA. |
9 pag...9 pages, 16 figures, 3 table |
| Out-of-Sight Embodied Agents: Multimodal Tracking, Sensor Fusion, and Trajectory Forecasting | 2026-03-26 | ShowTrajectory prediction is a fundamental problem in computer vision, vision-language-action models, world models, and autonomous systems, with broad impact on autonomous driving, robotics, and surveillance. However, most existing methods assume complete and clean observations, and therefore do not adequately handle out-of-sight agents or noisy sensing signals caused by limited camera coverage, occlusions, and the absence of ground-truth denoised trajectories. These challenges raise safety concerns and reduce robustness in real-world deployment. In this extended study, we introduce major improvements to Out-of-Sight Trajectory (OST), a task for predicting noise-free visual trajectories of out-of-sight objects from noisy sensor observations. Building on our prior work, we expand Out-of-Sight Trajectory Prediction (OOSTraj) from pedestrians to both pedestrians and vehicles, increasing its relevance to autonomous driving, robotics, and surveillance. Our improved Vision-Positioning Denoising Module exploits camera calibration to establish vision-position correspondence, mitigating the lack of direct visual cues and enabling effective unsupervised denoising of noisy sensor signals. Extensive experiments on the Vi-Fi and JRDB datasets show that our method achieves state-of-the-art results for both trajectory denoising and trajectory prediction, with clear gains over prior baselines. We also compare with classical denoising methods, including Kalman filtering, and adapt recent trajectory prediction models to this setting, establishing a stronger benchmark. To the best of our knowledge, this is the first work to use vision-positioning projection to denoise noisy sensor trajectories of out-of-sight agents, opening new directions for future research. |
Publi...Published in IEEE Transactions on Pattern Analysis and Machine Intelligence (Early Access), pp. 1-14, March 23, 2026 |
| Drive My Way: Preference Alignment of Vision-Language-Action Model for Personalized Driving | 2026-03-26 | ShowHuman driving behavior is inherently personal, which is shaped by long-term habits and influenced by short-term intentions. Individuals differ in how they accelerate, brake, merge, yield, and overtake across diverse situations. However, existing end-to-end autonomous driving systems either optimize for generic objectives or rely on fixed driving modes, lacking the ability to adapt to individual preferences or interpret natural language intent. To address this gap, we propose Drive My Way (DMW), a personalized Vision-Language-Action (VLA) driving framework that aligns with users' long-term driving habits and adapts to real-time user instructions. DMW learns a user embedding from our personalized driving dataset collected across multiple real drivers and conditions the policy on this embedding during planning, while natural language instructions provide additional short-term guidance. Closed-loop evaluation on the Bench2Drive benchmark demonstrates that DMW improves style instruction adaptation, and user studies show that its generated behaviors are recognizable as each driver's own style, highlighting personalization as a key capability for human-centered autonomous driving. Our data and code are available at https://dmw-cvpr.github.io/. |
IEEE/...IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2026); Project website: https://dmw-cvpr.github.io/ |
| Bridging Language and Action: A Survey of Language-Conditioned Robot Manipulation | 2026-03-26 | ShowLanguage-conditioned robot manipulation is an emerging field aimed at enabling seamless communication and cooperation between humans and robotic agents by teaching robots to comprehend and execute instructions conveyed in natural language. This interdisciplinary area integrates scene understanding, language processing, and policy learning to bridge the gap between human instructions and robot actions. In this comprehensive survey, we systematically explore recent advancements in language-conditioned robot manipulation. We categorize existing methods based on the primary ways language is integrated into the robot system, namely language for state evaluation, language as a policy condition, language for cognitive planning and reasoning, and language in unified vision-language-action models. Specifically, we further analyze state-of-the-art techniques from five axes of action granularity, data and supervision regimes, system cost and latency, environments and evaluations, and cross-modal task specification. Additionally, we highlight the key debates in the field. Finally, we discuss open challenges and future research directions, focusing on potentially enhancing generalization capabilities and addressing safety issues in language-conditioned robot manipulators. |
|
| LILAC: Language-Conditioned Object-Centric Optical Flow for Open-Loop Trajectory Generation | 2026-03-26 | ShowWe address language-conditioned robotic manipulation using flow-based trajectory generation, which enables training on human and web videos of object manipulation and requires only minimal embodiment-specific data. This task is challenging, as object trajectory generation from pre-manipulation images and natural language instructions requires appropriate instruction-flow alignment. To tackle this challenge, we propose the flow-based Language Instruction-guided open-Loop ACtion generator (LILAC). This flow-based Vision-Language-Action model (VLA) generates object-centric 2D optical flow from an RGB image and a natural language instruction, and converts the flow into a 6-DoF manipulator trajectory. LILAC incorporates two key components: Semantic Alignment Loss, which strengthens language conditioning to generate instruction-aligned optical flow, and Prompt-Conditioned Cross-Modal Adapter, which aligns learned visual prompts with image and text features to provide rich cues for flow generation. Experimentally, our method outperformed existing approaches in generated flow quality across multiple benchmarks. Furthermore, in physical object manipulation experiments using free-form instructions, LILAC demonstrated a superior task success rate compared to existing methods. The project page is available at https://lilac-75srg.kinsta.page/. |
Accep...Accepted to IEEE RA-L |
| LaMP: Learning Vision-Language-Action Policies with 3D Scene Flow as Latent Motion Prior | 2026-03-26 | ShowWe introduce \textbf{LaMP}, a dual-expert Vision-Language-Action framework that embeds dense 3D scene flow as a latent motion prior for robotic manipulation. Existing VLA models regress actions directly from 2D semantic visual features, forcing them to learn complex 3D physical interactions implicitly. This implicit learning strategy degrades under unfamiliar spatial dynamics. LaMP addresses this limitation by aligning a flow-matching \emph{Motion Expert} with a policy-predicting \emph{Action Expert} through gated cross-attention. Specifically, the Motion Expert generates a one-step partially denoised 3D scene flow, and its hidden states condition the Action Expert without full multi-step reconstruction. We evaluate LaMP on the LIBERO, LIBERO-Plus, and SimplerEnv-WidowX simulation benchmarks as well as real-world experiments. LaMP consistently outperforms evaluated VLA baselines across LIBERO, LIBERO-Plus, and SimplerEnv-WidowX benchmarks, achieving the highest reported average success rates under the same training budgets. On LIBERO-Plus OOD perturbations, LaMP shows improved robustness with an average 9.7% gain over the strongest prior baseline. Our project page is available at https://summerwxk.github.io/lamp-project-page/. |
|
| Diagnose, Correct, and Learn from Manipulation Failures via Visual Symbols | 2026-03-26 | ShowVision-Language-Action (VLA) models have recently achieved remarkable progress in robotic manipulation, yet they remain limited in failure diagnosis and learning from failures. Additionally, existing failure datasets are mostly generated programmatically in simulation, which limits their generalization to the real world. In light of these, we introduce ViFailback, a framework designed to diagnose robotic manipulation failures and provide both textual and visual correction guidance. Our framework utilizes explicit visual symbols to enhance annotation efficiency. We further release the ViFailback dataset, a large-scale collection of 58,126 Visual Question Answering (VQA) pairs along with their corresponding 5,202 real-world manipulation trajectories. Based on the dataset, we establish ViFailback-Bench, a benchmark of 11 fine-grained VQA tasks designed to assess the failure diagnosis and correction abilities of Vision-Language Models (VLMs), featuring ViFailback-Bench Lite for closed-ended and ViFailback-Bench Hard for open-ended evaluation. To demonstrate the effectiveness of our framework, we built the ViFailback-8B VLM, which not only achieves significant overall performance improvement on ViFailback-Bench but also generates visual symbols for corrective action guidance. Finally, by integrating ViFailback-8B with a VLA model, we conduct real-world robotic experiments demonstrating its ability to assist the VLA model in recovering from failures. Project Website: https://x1nyuzhou.github.io/vifailback.github.io/ |
Accep...Accepted by CVPR 2026. Project Website: https://x1nyuzhou.github.io/vifailback.github.io/ |
| ETA-VLA: Efficient Token Adaptation via Temporal Fusion and Intra-LLM Sparsification for Vision-Language-Action Models | 2026-03-26 | ShowThe integration of Vision-Language-Action (VLA) models into autonomous driving systems offers a unified framework for interpreting complex scenes and executing control commands. However, the necessity to incorporate historical multi-view frames for accurate temporal reasoning imposes a severe computational burden, primarily driven by the quadratic complexity of self-attention mechanisms in Large Language Models (LLMs). To alleviate this bottleneck, we propose ETA-VLA, an Efficient Token Adaptation framework for VLA models. ETA-VLA processes the past |
|
| $π$, But Make It Fly: Physics-Guided Transfer of VLA Models to Aerial Manipulation | 2026-03-26 | ShowVision-Language-Action (VLA) models such as |
|
| Beyond Attention Magnitude: Leveraging Inter-layer Rank Consistency for Efficient Vision-Language-Action Models | 2026-03-26 | ShowVision-Language-Action (VLA) models excel in robotic manipulation but suffer from significant inference latency due to processing dense visual tokens. Existing token reduction methods predominantly rely on attention magnitude as a static selection. In this work, we challenge this assumption, revealing that high-attention tokens are task-dependent and can even degrade policy performance. To address this, we introduce \textbf{TIES} (\textbf{T}au-guided \textbf{I}nter-layer \textbf{E}fficient \textbf{S}election), a dynamic framework guided by inter-layer token ranking consistency. By adaptively balancing attention magnitude with ranking consistency, TIES ensures robust token selection without requiring additional training. On the CogACT + SIMPLER benchmark, TIES improves average success rates by 6% while reducing token usage by 78%, and demonstrate strong generalization across diverse decoders and benchmarks. |
10 pa...10 pages, 7 figures, preprint |
| SABER: A Stealthy Agentic Black-Box Attack Framework for Vision-Language-Action Models | 2026-03-26 | ShowVision-language-action (VLA) models enable robots to follow natural-language instructions grounded in visual observations, but the instruction channel also introduces a critical vulnerability: small textual perturbations can alter downstream robot behavior. Systematic robustness evaluation therefore requires a black-box attacker that can generate minimal yet effective instruction edits across diverse VLA models. To this end, we present SABER, an agent-centric approach for automatically generating instruction-based adversarial attacks on VLA models under bounded edit budgets. SABER uses a GRPO-trained ReAct attacker to generate small, plausible adversarial instruction edits using character-, token-, and prompt-level tools under a bounded edit budget that induces targeted behavioral degradation, including task failure, unnecessarily long execution, and increased constraint violations. On the LIBERO benchmark across six state-of-the-art VLA models, SABER reduces task success by 20.6%, increases action-sequence length by 55%, and raises constraint violations by 33%, while requiring 21.1% fewer tool calls and 54.7% fewer character edits than strong GPT-based baselines. These results show that small, plausible instruction edits are sufficient to substantially degrade robot execution, and that an agentic black-box pipeline offers a practical, scalable, and adaptive approach for red-teaming robotic foundation models. |
|
| TAG: Target-Agnostic Guidance for Stable Object-Centric Inference in Vision-Language-Action Models | 2026-03-25 | ShowVision--Language--Action (VLA) policies have shown strong progress in mapping language instructions and visual observations to robotic actions, yet their reliability degrades in cluttered scenes with distractors. By analyzing failure cases, we find that many errors do not arise from infeasible motions, but from instance-level grounding failures: the policy often produces a plausible grasp trajectory that lands slightly off-target or even on the wrong object instance. To address this issue, we propose TAG (Target-Agnostic Guidance), a simple inference-time guidance mechanism that explicitly reduces distractor- and appearance-induced bias in VLA policies. Inspired by classifier-free guidance (CFG), TAG contrasts policy predictions under the original observation and an object-erased observation, and uses their difference as a residual steering signal that strengthens the influence of object evidence in the decision process. TAG does not require modifying the policy architecture and can be integrated with existing VLA policies with minimal training and inference changes. We evaluate TAG on standard manipulation benchmarks, including LIBERO, LIBERO-Plus, and VLABench, where it consistently improves robustness under clutter and reduces near-miss and wrong-object executions. |
|
| 3D-Mix for VLA: A Plug-and-Play Module for Integrating VGGT-based 3D Information into Vision-Language-Action Models | 2026-03-25 | ShowVision-Language-Action (VLA) models leverage Multimodal Large Language Models (MLLMs) for robotic control, but recent studies reveal that MLLMs exhibit limited spatial intelligence due to training predominantly on 2D data, resulting in inadequate 3D perception for manipulation tasks. While recent approaches incorporate specialized 3D vision models such as VGGT to enhance spatial understanding, they employ diverse integration mechanisms without systematic investigation, leaving the optimal fusion strategy unclear. We conduct a comprehensive pilot study comparing nine VGGT integration schemes on standardized benchmarks and find that semantic-conditioned gated fusion, which adaptively balances 2D semantic and 3D geometric features based on task context, achieved the strongest performance among all nine evaluated fusion schemes in our pilot study. We present 3D-Mix, a plug-and-play module that integrates into diverse VLA architectures (GR00T-style and |
13 pages |
| ACG: Action Coherence Guidance for Flow-based Vision-Language-Action models | 2026-03-25 | ShowDiffusion and flow matching models have emerged as powerful robot policies, enabling Vision-Language-Action (VLA) models to generalize across diverse scenes and instructions. Yet, when trained via imitation learning, their high generative capacity makes them sensitive to noise in human demonstrations: jerks, pauses, and jitter which reduce action coherence. Reduced action coherence causes instability and trajectory drift during deployment, failures that are catastrophic in fine-grained manipulation where precision is crucial. In this paper, we present Action Coherence Guidance (ACG) for VLA models, a training-free test-time guidance algorithm that improves action coherence and thereby yields performance gains. Evaluated on RoboCasa, DexMimicGen, and real-world SO-101 tasks, ACG consistently improves action coherence and boosts success rates across diverse manipulation tasks. Code and project page are available at https://github.com/DAVIAN-Robotics/ACG and https://DAVIAN-Robotics.github.io/ACG , respectively. |
Accep...Accepted to ICRA 2026 |
| E0: Enhancing Generalization and Fine-Grained Control in VLA Models via Tweedie Discrete Diffusion | 2026-03-25 | ShowVision-Language-Action (VLA) models offer a unified framework for robotic manipulation by integrating visual perception, language understanding, and control generation. However, existing VLA systems still struggle to generalize across diverse tasks, scenes, and camera viewpoints, and often produce coarse or unstable actions. We argue that these limitations are closely tied to the structural properties of actions in VLA settings, including the inherent multi-peaked nature of action distributions, the token-based symbolic reasoning of pretrained VLM/VLA backbones, and the effective finite resolution imposed by real-world robotic control. Motivated by these properties, we introduce E0, a tweedie discrete diffusion framework that formulates action generation as iterative denoising over quantized action tokens. By operating in a discrete action space with a principled diffusion process, E0 naturally aligns with token-based reasoning, supports fine-grained yet executable action control, and avoids the distributional mismatch of masking-based discrete diffusion. We further introduce a spherical viewpoint perturbation augmentation to enhance robustness to camera shifts without additional data. Experiments on LIBERO, VLABench, ManiSkill, and a real-world Franka arm demonstrate that E0 achieves state-of-the-art performance across 14 diverse environments, outperforming strong baselines by 10.7% on average. |
|
| Unified Diffusion VLA: Vision-Language-Action Model via Joint Discrete Denoising Diffusion Process | 2026-03-25 | ShowVision-language-action (VLA) models aim to understand natural language instructions and visual observations and to execute corresponding actions as an embodied agent. Recent work integrates future images into the understanding-acting loop, yielding unified VLAs that jointly understand, generate, and act -- reading text and images and producing future images and actions. However, these models either rely on external experts for modality unification or treat image generation and action prediction as separate processes, limiting the benefits of direct synergy between these tasks. Our core philosophy is to optimize generation and action jointly through a synchronous denoising process, where the iterative refinement enables actions to evolve from initialization, under constant and sufficient visual guidance. We ground this philosophy in our proposed Unified Diffusion VLA and Joint Discrete Denoising Diffusion Process (JD3P), which is a joint diffusion process that integrates multiple modalities into a single denoising trajectory to serve as the key mechanism enabling understanding, generation, and acting to be intrinsically synergistic. Our model and theory are built on a unified tokenized space of all modalities and a hybrid attention mechanism. We further propose a two-stage training pipeline and several inference-time techniques that optimize performance and efficiency. Our approach achieves state-of-the-art performance on benchmarks such as CALVIN, LIBERO, and SimplerEnv with 4$\times$ faster inference than autoregressive methods, and we demonstrate its effectiveness through in-depth analysis and real-world evaluations. Our project page is available at https://irpn-eai.github.io/UD-VLA.github.io/. |
|
| Xiaomi-Robotics-0: An Open-Sourced Vision-Language-Action Model with Real-Time Execution | 2026-03-25 | ShowIn this report, we introduce Xiaomi-Robotics-0, an advanced vision-language-action (VLA) model optimized for high performance and fast and smooth real-time execution. The key to our method lies in a carefully designed training recipe and deployment strategy. Xiaomi-Robotics-0 is first pre-trained on large-scale cross-embodiment robot trajectories and vision-language data, endowing it with broad and generalizable action-generation capabilities while avoiding catastrophic forgetting of the visual-semantic knowledge of the underlying pre-trained VLM. During post-training, we propose several techniques for training the VLA model for asynchronous execution to address the inference latency during real-robot rollouts. During deployment, we carefully align the timesteps of consecutive predicted action chunks to ensure continuous and seamless real-time rollouts. We evaluate Xiaomi-Robotics-0 extensively in simulation benchmarks and on two challenging real-robot tasks that require precise and dexterous bimanual manipulation. Results show that our method achieves state-of-the-art performance across all simulation benchmarks. Moreover, Xiaomi-Robotics-0 can roll out fast and smoothly on real robots using a consumer-grade GPU, achieving high success rates and throughput on both real-robot tasks. To facilitate future research, code and model checkpoints are open-sourced at https://xiaomi-robotics-0.github.io |
Proje...Project page: https://xiaomi-robotics-0.github.io |
| Red-Teaming Vision-Language-Action Models via Quality Diversity Prompt Generation for Robust Robot Policies | 2026-03-24 | ShowVision-Language-Action (VLA) models have significant potential to enable general-purpose robotic systems for a range of vision-language tasks. However, the performance of VLA-based robots is highly sensitive to the precise wording of language instructions, and it remains difficult to predict when such robots will fail. To improve the robustness of VLAs to different wordings, we present Q-DIG (Quality Diversity for Diverse Instruction Generation), which performs red-teaming by scalably identifying diverse natural language task descriptions that induce failures while remaining task-relevant. Q-DIG integrates Quality Diversity (QD) techniques with Vision-Language Models (VLMs) to generate a broad spectrum of adversarial instructions that expose meaningful vulnerabilities in VLA behavior. Our results across multiple simulation benchmarks show that Q-DIG finds more diverse and meaningful failure modes compared to baseline methods, and that fine-tuning VLAs on the generated instructions improves task success rates. Furthermore, results from a user study highlight that Q-DIG generates prompts judged to be more natural and human-like than those from baselines. Finally, real-world evaluations of Q-DIG prompts show results consistent with simulation, and fine-tuning VLAs on the generated prompts further success rates on unseen instructions. Together, these findings suggest that Q-DIG is a promising approach for identifying vulnerabilities and improving the robustness of VLA-based robots. Our anonymous project website is at qdigvla.github.io. |
|
| TacVLA: Contact-Aware Tactile Fusion for Robust Vision-Language-Action Manipulation | 2026-03-24 | ShowVision-Language-Action (VLA) models have demonstrated significant advantages in robotic manipulation. However, their reliance on vision and language often leads to suboptimal performance in tasks involving visual occlusion, fine-grained manipulation, and physical contact. To address these challenges, we propose TacVLA, a fine-tuned VLA model by incorporating tactile modalities into the transformer-based policy to enhance fine-grained manipulation capabilities. Specifically, we introduce a contact-aware gating mechanism that selectively activates tactile tokens only when contact is detected, enabling adaptive multimodal fusion while avoiding irrelevant tactile interference. The fused visual, language, and tactile tokens are jointly processed within the transformer architecture to strengthen cross-modal grounding during contact-rich interaction. Extensive experiments on constraint-locked disassembly, in-box picking and robustness evaluations demonstrate that our model outperforms baselines, improving the performance by averaging 20% success rate in disassembly, 60% in in-box picking and 2.1x improvement in scenarios with visual occlusion. Videos are available at https://sites.google.com/view/tacvla and code will be released. |
9 pages, 7 figures |
| Gaze-Regularized Vision-Language-Action Models for Robotic Manipulation | 2026-03-24 | ShowDespite advances in Vision-Language-Action (VLA) models, robotic manipulation struggles with fine-grained tasks because current models lack mechanisms for active visual attention allocation. Human gaze naturally encodes intent, planning, and execution patterns -- offering a powerful supervisory signal for guiding robot perception. We introduce a gaze-regularized training framework that aligns VLA models' internal attention with human visual patterns without architectural modifications or inference-time overhead. Our method transforms temporally aggregated gaze heatmaps into patch-level distributions and regularizes the transformer's attention through KL divergence, creating an inductive bias toward task-relevant features while preserving deployment efficiency. When integrated into existing VLA architectures, our approach yields 4-12% improvements across manipulation benchmarks. The gaze-regularized models reach equivalent performance with fewer training steps and maintain robustness under lighting variations and sensor noise. Beyond performance metrics, the learned attention patterns produce interpretable visualizations that mirror human strategies, enhancing trust in robotic systems. Moreover, our framework requires no eye-tracking equipment and applies directly to existing datasets. These results demonstrate that human perceptual priors can significantly accelerate robot learning while improving both task performance and system interpretability. |
|
| TRAP: Hijacking VLA CoT-Reasoning via Adversarial Patches | 2026-03-24 | ShowBy integrating Chain-of-Thought(CoT) reasoning, Vision-Language-Action (VLA) models have demonstrated strong capabilities in robotic manipulation, particularly by improving generalization and interpretability. However, the security of CoT-based reasoning mechanisms remains largely unexplored. In this paper, we show that CoT reasoning introduces a novel attack vector for targeted control hijacking--for example, causing a robot to mistakenly deliver a knife to a person instead of an apple--without modifying the user's instruction. We first provide empirical evidence that CoT strongly governs action generation, even when it is semantically misaligned with the input instructions. Building on this observation, we propose TRAP, the first targeted adversarial attack framework for CoT-reasoning VLA models. TRAP uses an adversarial patch (e.g., a coaster placed on the table) to corrupt intermediate CoT reasoning and hijack the VLA's output. By optimizing the CoT adversarial loss, TRAP induces specific and adversary-defined behaviors. Extensive evaluations across 3 mainstream VLA architectures and 3 CoT reasoning paradigms validate the effectiveness of TRAP. Notably, we implemented the patch by printing it on paper in a real-world setting. Our findings highlight the urgent need to secure CoT reasoning in VLA systems. |
|
| VLA-IAP: Training-Free Visual Token Pruning via Interaction Alignment for Vision-Language-Action Models | 2026-03-24 | ShowVision-Language-Action (VLA) models have rapidly advanced embodied intelligence, enabling robots to execute complex, instruction-driven tasks. However, as model capacity and visual context length grow, the inference cost of VLA systems becomes a major bottleneck for real-world deployment on resource-constrained platforms. Existing visual token pruning methods mainly rely on semantic saliency or simple temporal cues, overlooking the continuous physical interaction, a fundamental property of VLA tasks. Consequently, current approaches often prune visually sparse yet structurally critical regions that support manipulation, leading to unstable behavior during early task phases. To overcome this, we propose a shift toward an explicit Interaction-First paradigm. Our proposed \textbf{training-free} method, VLA-IAP (Interaction-Aligned Pruning), introduces a geometric prior mechanism to preserve structural anchors and a dynamic scheduling strategy that adapts pruning intensity based on semantic-motion alignment. This enables a conservative-to-aggressive transition, ensuring robustness during early uncertainty and efficiency once interaction is locked. Extensive experiments show that VLA-IAP achieves a \textbf{97.8% success rate} with a \textbf{$1.25\times$ speedup} on the LIBERO benchmark, and up to \textbf{$1.54\times$ speedup} while maintaining performance \textbf{comparable to the unpruned backbone}. Moreover, the method demonstrates superior and consistent performance across multiple model architectures and three different simulation environments, as well as a real robot platform, validating its strong generalization capability and practical applicability. Our project website is: \href{https://chengjt1999.github.io/VLA-IAP.github.io/}{VLA-IAP.com}. |
27 pages, 8 figures |
| Agile-VLA: Few-Shot Industrial Pose Rectification via Implicit Affordance Anchoring | 2026-03-24 | ShowDeploying Vision-Language-Action (VLA) models on resource-constrained edge platforms encounters a fundamental conflict between high-latency semantic inference and the high-frequency control required for dynamic manipulation. To address the challenge, this paper presents Agile-VLA, a hierarchical framework designed for industrial pose reorientation tasks on edge devices such as the NVIDIA Jetson Orin Nano. The core innovation is an Implicit Affordance Anchoring mechanism that directly maps geometric visual cues, specifically centroid and rim keypoint anchors, into structured parametric action primitives, thereby substantially reducing reliance on high-latency semantic inference during closed-loop control. By decoupling perception (10 Hz) from control (50 Hz) via an asynchronous dual-stream architecture, the system effectively mitigates the frequency mismatch inherent in edge-based robot learning. Experimental results on a standard 6-DoF manipulator demonstrate that Agile-VLA achieves robust rectification of complex, irregular workpieces using only 5-shot demonstrations through extrinsic dexterity. |
8 pag...8 pages. Submitted to IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2026 |
| Grounding Sim-to-Real Generalization in Dexterous Manipulation: An Empirical Study with Vision-Language-Action Models | 2026-03-24 | ShowLearning a generalist control policy for dexterous manipulation typically relies on large-scale datasets. Given the high cost of real-world data collection, a practical alternative is to generate synthetic data through simulation. However, the resulting synthetic data often exhibits a significant gap from real-world distributions. While many prior studies have proposed algorithms to bridge the Sim-to-Real discrepancy, there remains a lack of principled research that grounds these methods in real-world manipulation tasks, particularly their performance on generalist policies such as Vision-Language-Action (VLA) models. In this study, we empirically examine the primary determinants of Sim-to-Real generalization across four dimensions: multi-level domain randomization, photorealistic rendering, physics-realistic modeling, and reinforcement learning updates. To support this study, we design a comprehensive evaluation protocol to quantify the real-world performance of manipulation tasks. The protocol accounts for key variations in background, lighting, distractors, object types, and spatial features. Through experiments involving over 10k real-world trials, we derive critical insights into Sim-to-Real transfer. To inform and advance future studies, we release both the robotic platforms and the evaluation protocol for public access to facilitate independent verification, thereby establishing a realistic and standardized benchmark for dexterous manipulation policies. |
|
| CATNAV: Cached Vision-Language Traversability for Efficient Zero-Shot Robot Navigation | 2026-03-24 | ShowNavigating unstructured environments requires assessing traversal risk relative to a robot's physical capabilities, a challenge that varies across embodiments. We present CATNAV, a cost-aware traversability navigation framework that leverages multimodal LLMs for zero-shot, embodiment-aware costmap generation without task-specific training. We introduce a visuosemantic caching mechanism that detects scene novelty and reuses prior risk assessments for semantically similar frames, reducing online VLM queries by 85.7%. Furthermore, we introduce a VLM-based trajectory selection module that evaluates proposals through visual reasoning to choose the safest path given behavioral constraints. We evaluate CATNAV on a quadruped robot across indoor and outdoor unstructured environments, comparing against state-of-the-art vision-language-action baselines. Across five navigation tasks, CATNAV achieves 10 percentage point higher average goal-reaching rate and 33% fewer behavioral constraint violations. |
8 pages, 6 figures |
| SG-VLA: Learning Spatially-Grounded Vision-Language-Action Models for Mobile Manipulation | 2026-03-24 | ShowVision-Language-Action (VLA) models show promise for robotic control, yet performance in complex household environments remains sub-optimal. Mobile manipulation requires reasoning about global scene layout, fine-grained geometry, and high-dimensional continuous actions, making standard imitation learning insufficient. We introduce a framework for learning spatially-grounded VLA models that strengthens perception and representation through auxiliary task co-training and multi-modal input enhancement. Our method addresses the challenge of controlling a 13-dimensional action space involving coordinated base motion, arm articulation, and gripper actuation. To enrich spatial understanding, the model incorporates multi-view RGB observations, depth cues, and short temporal history, providing perspectives of both global scene structure and local manipulation context. To improve representation quality, we co-train auxiliary decoders that reconstruct interpretable intermediate signals - including global robot position, joint configurations, grasp affordances, target-object relative pose, and segmentation masks - from shared visual-language features. These objectives provide dense supervision that encourages the backbone to develop spatially grounded, manipulation-aware latent representations. Through extensive evaluation on home rearrangement tasks, our approach achieves consistent improvements across picking, placing, opening, and closing operations, substantially outperforming direct imitation learning. Our findings suggest that spatial grounding through auxiliary and multi-modal learning provides a strong direction for scaling VLA models toward general-purpose domestic robots. |
|
| Point What You Mean: Visually Grounded Instruction Policy | 2026-03-24 | ShowVision-Language-Action (VLA) models align vision and language with embodied control, but their object referring ability remains limited when relying solely on text prompt, especially in cluttered or out-of-distribution (OOD) scenes. In this study, we introduce the Point-VLA, a plug-and-play policy that augments language instructions with explicit visual cues (e.g., bounding boxes) to resolve referential ambiguity and enable precise object-level grounding. To efficiently scale visually grounded datasets, we further develop an automatic data annotation pipeline requiring minimal human effort. We evaluate Point-VLA on diverse real-world referring tasks and observe consistently stronger performance than text-only instruction VLAs, particularly in cluttered or unseen-object scenarios, with robust generalization. These results demonstrate that Point-VLA effectively resolves object referring ambiguity through pixel-level visual grounding, achieving more generalizable embodied control. |
|
| CaP-X: A Framework for Benchmarking and Improving Coding Agents for Robot Manipulation | 2026-03-23 | Show"Code-as-Policy" considers how executable code can complement data-intensive Vision-Language-Action (VLA) methods, yet their effectiveness as autonomous controllers for embodied manipulation remains underexplored. We present CaP-X, an open-access framework for systematically studying Code-as-Policy agents in robot manipulation. At its core is CaP-Gym, an interactive environment in which agents control robots by synthesizing and executing programs that compose perception and control primitives. Building on this foundation, CaP-Bench evaluates frontier language and vision-language models across varying levels of abstraction, interaction, and perceptual grounding. Across 12 models, CaP-Bench reveals a consistent trend: performance improves with human-crafted abstractions but degrades as these priors are removed, exposing a dependence on designer scaffolding. At the same time, we observe that this gap can be mitigated through scaling agentic test-time computation--through multi-turn interaction, structured execution feedback, visual differencing, automatic skill synthesis, and ensembled reasoning--substantially improves robustness even when agents operate over low-level primitives. These findings allow us to derive CaP-Agent0, a training-free framework that recovers human-level reliability on several manipulation tasks in simulation and on real embodiments. We further introduce CaP-RL, showing reinforcement learning with verifiable rewards improves success rates and transfers from sim2real with minimal gap. Together, CaP-X provides a principled, open-access platform for advancing embodied coding agents. |
|
| DualCoT-VLA: Visual-Linguistic Chain of Thought via Parallel Reasoning for Vision-Language-Action Models | 2026-03-23 | ShowVision-Language-Action (VLA) models map visual observations and language instructions directly to robotic actions. While effective for simple tasks, standard VLA models often struggle with complex, multi-step tasks requiring logical planning, as well as precise manipulations demanding fine-grained spatial perception. Recent efforts have incorporated Chain-of-Thought (CoT) reasoning to endow VLA models with a ``thinking before acting'' capability. However, current CoT-based VLA models face two critical limitations: 1) an inability to simultaneously capture low-level visual details and high-level logical planning due to their reliance on isolated, single-modal CoT; 2) high inference latency with compounding errors caused by step-by-step autoregressive decoding. To address these limitations, we propose DualCoT-VLA, a visual-linguistic CoT method for VLA models with a parallel reasoning mechanism. To achieve comprehensive multi-modal reasoning, our method integrates a visual CoT for low-level spatial understanding and a linguistic CoT for high-level task planning. Furthermore, to overcome the latency bottleneck, we introduce a parallel CoT mechanism that incorporates two sets of learnable query tokens, shifting autoregressive reasoning to single-step forward reasoning. Extensive experiments demonstrate that our DualCoT-VLA achieves state-of-the-art performance on the LIBERO and RoboCasa GR1 benchmarks, as well as in real-world platforms. |
|
| UniDex: A Robot Foundation Suite for Universal Dexterous Hand Control from Egocentric Human Videos | 2026-03-23 | ShowDexterous manipulation remains challenging due to the cost of collecting real-robot teleoperation data, the heterogeneity of hand embodiments, and the high dimensionality of control. We present UniDex, a robot foundation suite that couples a large-scale robot-centric dataset with a unified vision-language-action (VLA) policy and a practical human-data capture setup for universal dexterous hand control. First, we construct UniDex-Dataset, a robot-centric dataset over 50K trajectories across eight dexterous hands (6--24 DoFs), derived from egocentric human video datasets. To transform human data into robot-executable trajectories, we employ a human-in-the-loop retargeting procedure to align fingertip trajectories while preserving plausible hand-object contacts, and we operate on explicit 3D pointclouds with human hands masked to narrow kinematic and visual gaps. Second, we introduce the Function-Actuator-Aligned Space (FAAS), a unified action space that maps functionally similar actuators to shared coordinates, enabling cross-hand transfer. Leveraging FAAS as the action parameterization, we train UniDex-VLA, a 3D VLA policy pretrained on UniDex-Dataset and finetuned with task demonstrations. In addition, we build UniDex-Cap, a simple portable capture setup that records synchronized RGB-D streams and human hand poses and converts them into robot-executable trajectories to enable human-robot data co-training that reduces reliance on costly robot demonstrations. On challenging tool-use tasks across two different hands, UniDex-VLA achieves 81% average task progress and outperforms prior VLA baselines by a large margin, while exhibiting strong spatial, object, and zero-shot cross-hand generalization. Together, UniDex-Dataset, UniDex-VLA, and UniDex-Cap provide a scalable foundation suite for universal dexterous manipulation. |
Accep...Accepted by CVPR 2026 |
| Foundation Models for Trajectory Planning in Autonomous Driving: A Review of Progress and Open Challenges | 2026-03-23 | ShowThe emergence of multi-modal foundation models has markedly transformed the technology for autonomous driving, shifting away from conventional and mostly hand-crafted design choices towards unified, foundation-model-based approaches, capable of directly inferring motion trajectories from raw sensory inputs. This new class of methods can also incorporate natural language as an additional modality, with Vision-Language-Action (VLA) models serving as a representative example. In this review, we provide a comprehensive examination of such methods through a unifying taxonomy to critically evaluate their architectural design choices, methodological strengths, and their inherent capabilities and limitations. Our survey covers 37 recently proposed approaches that span the landscape of trajectory planning with foundation models. Furthermore, we assess these approaches with respect to the openness of their source code and datasets, offering valuable information to practitioners and researchers. We provide an accompanying webpage that catalogues the methods based on our taxonomy, available at: https://github.com/fiveai/FMs-for-driving-trajectories |
Accep...Accepted to TMLR (Survey Certification) |
| VP-VLA: Visual Prompting as an Interface for Vision-Language-Action Models | 2026-03-23 | ShowVision-Language-Action (VLA) models typically map visual observations and linguistic instructions directly to robotic control signals. This "black-box" mapping forces a single forward pass to simultaneously handle instruction interpretation, spatial grounding, and low-level control, often leading to poor spatial precision and limited robustness in out-of-distribution scenarios. To address these limitations, we propose VP-VLA, a dual-system framework that decouples high-level reasoning and low-level execution via a structured visual prompting interface. Specifically, a "System 2 Planner" decomposes complex instructions into sub-tasks and identifies relevant target objects and goal locations. These spatial anchors are then overlaid directly onto visual observations as structured visual prompts, such as crosshairs and bounding boxes. Guided by these prompts and enhanced by a novel auxiliary visual grounding objective during training, a "System 1 Controller" reliably generates precise low-level execution motions. Experiments on the Robocasa-GR1-Tabletop benchmark and SimplerEnv simulation demonstrate that VP-VLA improves success rates by 5% and 8.3%, surpassing competitive baselines including QwenOFT and GR00T-N1.6. |
Proje...Project page: https://visualprompt-vla.github.io/ |
| Concept-Based Dictionary Learning for Inference-Time Safety in Vision Language Action Models | 2026-03-23 | ShowVision Language Action (VLA) models close the perception action loop by translating multimodal instructions into executable behaviors, but this very capability magnifies safety risks: jailbreaks that merely yield toxic text in LLMs can trigger unsafe physical actions in embodied systems. Existing defenses alignment, filtering, or prompt hardening intervene too late or at the wrong modality, leaving fused representations exploitable. We introduce a concept based dictionary learning framework for inference time safety control. By learning sparse, interpretable dictionaries from hidden activations, our method identifies harmful concept directions and attenuates risky components when the estimated risk exceeds a threshold. Experiments on Libero-Harm, BadRobot, RoboPair, and IS-Bench show that our approach achieves state-of-the-art defense performance, cutting attack success rates by over 70% while maintaining task success. Crucially, the framework is plug-in and model-agnostic, requiring no retraining and integrating seamlessly with diverse VLAs. To our knowledge, this is the first inference time concept based safety method for embodied systems, advancing both interpretability and safe deployment of VLA models. |
|
| AI Token Futures Market: Commoditization of Compute and Derivatives Contract Design | 2026-03-23 | ShowAs large language models (LLMs) and vision-language-action models (VLAs) become widely deployed, the tokens consumed by AI inference are evolving into a new type of commodity. This paper systematically analyzes the commodity attributes of tokens, arguing for their transition from intelligent service outputs to compute infrastructure raw materials, and draws comparisons with established commodities such as electricity, carbon emission allowances, and bandwidth. Building on the historical experience of electricity futures markets and the theory of commodity financialization, we propose a complete design for standardized token futures contracts, including the definition of a Standard Inference Token (SIT), contract specifications, settlement mechanisms, margin systems, and market-maker regimes. By constructing a mean-reverting jump-diffusion stochastic process model and conducting Monte Carlo simulations, we evaluate the hedging efficiency of the proposed futures contracts for application-layer enterprises. Simulation results show that, under an application-layer demand explosion scenario, token futures can reduce enterprise compute cost volatility by 62%-78%. We also explore the feasibility of GPU compute futures and discuss the regulatory framework for token futures markets, providing a theoretical foundation and practical roadmap for the financialization of compute resources. |
16 pa...16 pages, 7 figures, 3 tables |
| Fast-WAM: Do World Action Models Need Test-time Future Imagination? | 2026-03-23 | ShowWorld Action Models (WAMs) have emerged as a promising alternative to Vision-Language-Action (VLA) models for embodied control because they explicitly model how visual observations may evolve under action. Most existing WAMs follow an imagine-then-execute paradigm, incurring substantial test-time latency from iterative video denoising, yet it remains unclear whether explicit future imagination is actually necessary for strong action performance. In this paper, we ask whether WAMs need explicit future imagination at test time, or whether their benefit comes primarily from video modeling during training. We disentangle the role of video modeling during training from explicit future generation during inference by proposing \textbf{Fast-WAM}, a WAM architecture that retains video co-training during training but skips future prediction at test time. We further instantiate several Fast-WAM variants to enable a controlled comparison of these two factors. Across these variants, we find that Fast-WAM remains competitive with imagine-then-execute variants, while removing video co-training causes a much larger performance drop. Empirically, Fast-WAM achieves competitive results with state-of-the-art methods both on simulation benchmarks (LIBERO and RoboTwin) and real-world tasks, without embodied pretraining. It runs in real time with 190ms latency, over 4$\times$ faster than existing imagine-then-execute WAMs. These results suggest that the main value of video prediction in WAMs may lie in improving world representations during training rather than generating future observations at test time. Project page: https://yuantianyuan01.github.io/FastWAM/ |
|
| RoboAlign: Learning Test-Time Reasoning for Language-Action Alignment in Vision-Language-Action Models | 2026-03-22 | ShowImproving embodied reasoning in multimodal-large-language models (MLLMs) is essential for building vision-language-action models (VLAs) on top of them to readily translate multimodal understanding into low-level actions. Accordingly, recent work has explored enhancing embodied reasoning in MLLMs through supervision of vision-question-answering type. However, these approaches have been reported to result in unstable VLA performance, often yielding only marginal or even negative gains. In this paper, we propose a more systematic MLLM training framework RoboAlign that reliably improves VLA performance. Our key idea is to sample action tokens via zero-shot natural language reasoning and refines this reasoning using reinforcement learning (RL) to improve action accuracy. As a result, RoboAlign bridges the modality gap between language and low-level actions in MLLMs, and facilitate knowledge transfer from MLLM to VLA. To validate the effectiveness of RoboAlign, we train VLAs by adding a diffusion-based action head on top of an MLLM backbone and evaluate them on major robotics benchmarks. Remarkably, by performing RL-based alignment after SFT using less than 1% of the data, RoboAlign achieves performance improvements of 17.5%, 18.9%, and 106.6% over SFT baselines on LIBERO, CALVIN, and real-world environments, respectively. |
15 pa...15 pages, 7 figures, 9 Tables |
| RoboFAC: A Comprehensive Framework for Robotic Failure Analysis and Correction | 2026-03-22 | ShowVision-Language-Action (VLA) models have recently advanced robotic manipulation by translating natural-language instructions and visual observations into control actions. However, existing VLAs are primarily trained on successful expert demonstrations and lack structured supervision for failure diagnosis and recovery, limiting robustness in open-world scenarios. To address this limitation, we propose the Robotic Failure Analysis and Correction (RoboFAC) framework. We construct a large-scale failure-centric dataset comprising 9,440 erroneous manipulation trajectories and 78,623 QA pairs across 53 scenes in both simulation and real-world environments, with systematically categorized failure types. Leveraging this dataset, we develop a lightweight multimodal model specialized for task understanding, failure analysis, and failure correction, enabling efficient local deployment while remaining competitive with large proprietary models. Experimental results demonstrate that RoboFAC achieves a 34.1% higher failure analysis accuracy compared to GPT-4o. Furthermore, we integrated RoboFAC as an external supervisor in a real-world VLA control pipeline, yielding a 29.1% relative improvement across four tasks while significantly reducing latency relative to GPT-4o. These results demonstrate that RoboFAC enables systematic failure diagnosis and recovery, significantly enhancing VLA recovery capabilities. Our model and dataset are publicly available at https://github.com/MINT-SJTU/RoboFAC. |
| Title | Date | Abstract | Comment |
|---|---|---|---|
| Hierarchical Planning with Latent World Models | 2026-04-03 | ShowModel predictive control (MPC) with learned world models has emerged as a promising paradigm for embodied control, particularly for its ability to generalize zero-shot when deployed in new environments. However, learned world models often struggle with long-horizon control due to the accumulation of prediction errors and the exponentially growing search space. In this work, we address these challenges by learning latent world models at multiple temporal scales and performing hierarchical planning across these scales, enabling long-horizon reasoning while substantially reducing inference-time planning complexity. Our approach serves as a modular planning abstraction that applies across diverse latent world-model architectures and domains. We demonstrate that this hierarchical approach enables zero-shot control on real-world non-greedy robotic tasks, achieving a 70% success rate on pick-&-place using only a final goal specification, compared to 0% for a single-level world model. In addition, across physics-based simulated environments including push manipulation and maze navigation, hierarchical planning achieves higher success while requiring up to 4x less planning-time compute. |
|
| InCoder-32B-Thinking: Industrial Code World Model for Thinking | 2026-04-03 | ShowIndustrial software development across chip design, GPU optimization, and embedded systems lacks expert reasoning traces showing how engineers reason about hardware constraints and timing semantics. In this work, we propose InCoder-32B-Thinking, trained on the data from the Error-driven Chain-of-Thought (ECoT) synthesis framework with an industrial code world model (ICWM) to generate reasoning traces. Specifically, ECoT generates reasoning chains by synthesizing the thinking content from multi-turn dialogue with environmental error feedback, explicitly modeling the error-correction process. ICWM is trained on domain-specific execution traces from Verilog simulation, GPU profiling, etc., learns the causal dynamics of how code affects hardware behavior, and enables self-verification by predicting execution outcomes before actual compilation. All synthesized reasoning traces are validated through domain toolchains, creating training data matching the natural reasoning depth distribution of industrial tasks. Evaluation on 14 general (81.3% on LiveCodeBench v5) and 9 industrial benchmarks (84.0% in CAD-Coder and 38.0% on KernelBench) shows InCoder-32B-Thinking achieves top-tier open-source results across all domains.GPU Optimization |
|
| Learning Task-Invariant Properties via Dreamer: Enabling Efficient Policy Transfer for Quadruped Robots | 2026-04-03 | ShowAchieving quadruped robot locomotion across diverse and dynamic terrains presents significant challenges, primarily due to the discrepancies between simulation environments and real-world conditions. Traditional sim-to-real transfer methods often rely on manual feature design or costly real-world fine-tuning. To address these limitations, this paper proposes the DreamTIP framework, which incorporates Task-Invariant Properties learning within the Dreamer world model architecture to enhance sim-to-real transfer capabilities. Guided by large language models, DreamTIP identifies and leverages Task-Invariant Properties, such as contact stability and terrain clearance, which exhibit robustness to dynamic variations and strong transferability across tasks. These properties are integrated into the world model as auxiliary prediction targets, enabling the policy to learn representations that are insensitive to underlying dynamic changes. Furthermore, an efficient adaptation strategy is designed, employing a mixed replay buffer and regularization constraints to rapidly calibrate to real-world dynamics while effectively mitigating representation collapse and catastrophic forgetting. Extensive experiments on complex terrains, including Stair, Climb, Tilt, and Crawl, demonstrate that DreamTIP significantly outperforms state-of-the-art baselines in both simulated and real-world environments. Our method achieves an average performance improvement of 28.1% across eight distinct simulated transfer tasks. In the real-world Climb task, the baseline method achieved only a 10\ success rate, whereas our method attained a 100% success rate. These results indicate that incorporating Task-Invariant Properties into Dreamer learning offers a novel solution for achieving robust and transferable robot locomotion. |
Accep...Accepted by IEEE International Conference on Robotics and Automation (ICRA) 2026 |
| ExploreVLA: Dense World Modeling and Exploration for End-to-End Autonomous Driving | 2026-04-03 | ShowEnd-to-end autonomous driving models based on Vision-Language-Action (VLA) architectures have shown promising results by learning driving policies through behavior cloning on expert demonstrations. However, imitation learning inherently limits the model to replicating observed behaviors without exploring diverse driving strategies, leaving it brittle in novel or out-of-distribution scenarios. Reinforcement learning (RL) offers a natural remedy by enabling policy exploration beyond the expert distribution. Yet VLA models, typically trained on offline datasets, lack directly observable state transitions, necessitating a learned world model to anticipate action consequences. In this work, we propose a unified understanding-and-generation framework that leverages world modeling to simultaneously enable meaningful exploration and provide dense supervision. Specifically, we augment trajectory prediction with future RGB and depth image generation as dense world modeling objectives, requiring the model to learn fine-grained visual and geometric representations that substantially enrich the planning backbone. Beyond serving as a supervisory signal, the world model further acts as a source of intrinsic reward for policy exploration: its image prediction uncertainty naturally measures a trajectory's novelty relative to the training distribution, where high uncertainty indicates out-of-distribution scenarios that, if safe, represent valuable learning opportunities. We incorporate this exploration signal into a safety-gated reward and optimize the policy via Group Relative Policy Optimization (GRPO). Experiments on the NAVSIM and nuScenes benchmarks demonstrate the effectiveness of our approach, achieving a state-of-the-art PDMS score of 93.7 and an EPDMS of 88.8 on NAVSIM. The code and demo will be publicly available at https://zihaosheng.github.io/ExploreVLA/. |
The c...The code and demo will be publicly available at https://zihaosheng.github.io/ExploreVLA/ |
| VoxelCodeBench: Benchmarking 3D World Modeling Through Code Generation | 2026-04-02 | ShowEvaluating code generation models for 3D spatial reasoning requires executing generated code in realistic environments and assessing outputs beyond surface-level correctness. We introduce a platform VoxelCode, for analyzing code generation capabilities for 3D understanding and environment creation. Our platform integrates natural language task specification, API-driven code execution in Unreal Engine, and a unified evaluation pipeline supporting both automated metrics and human assessment. To demonstrate its utility, we construct VoxelCodeBench, a benchmark of voxel manipulation tasks spanning three reasoning dimensions: symbolic interpretation, geometric construction, and artistic composition. Evaluating leading code generation models, we find that producing executable code is far easier than producing spatially correct outputs, with geometric construction and multi-object composition proving particularly challenging. By open-sourcing our platform and benchmark, we provide the community with extensible infrastructure for developing new 3D code generation benchmarks and probing spatial reasoning in future models. |
|
| Learn2Fold: Structured Origami Generation with World Model Planning | 2026-04-02 | ShowThe ability to transform a flat sheet into a complex three-dimensional structure is a fundamental test of physical intelligence. Unlike cloth manipulation, origami is governed by strict geometric axioms and hard kinematic constraints, where a single invalid crease or collision can invalidate the entire folding sequence. As a result, origami demands long-horizon constructive reasoning that jointly satisfies precise physical laws and high-level semantic intent. Existing approaches fall into two disjoint paradigms: optimization-based methods enforce physical validity but require dense, precisely specified inputs, making them unsuitable for sparse natural language descriptions, while generative foundation models excel at semantic and perceptual synthesis yet fail to produce long-horizon, physics-consistent folding processes. Consequently, generating valid origami folding sequences directly from text remains an open challenge. To address this gap, we introduce Learn2Fold, a neuro-symbolic framework that formulates origami folding as conditional program induction over a crease-pattern graph. Our key insight is to decouple semantic proposal from physical verification. A large language model generates candidate folding programs from abstract text prompts, while a learned graph-structured world model serves as a differentiable surrogate simulator that predicts physical feasibility and failure modes before execution. Integrated within a lookahead planning loop, Learn2Fold enables robust generation of physically valid folding sequences for complex and out-of-distribution patterns, demonstrating that effective spatial intelligence arises from the synergy between symbolic reasoning and grounded physical simulation. |
9 pages, 6 figures |
| ActionParty: Multi-Subject Action Binding in Generative Video Games | 2026-04-02 | ShowRecent advances in video diffusion have enabled the development of "world models" capable of simulating interactive environments. However, these models are largely restricted to single-agent settings, failing to control multiple agents simultaneously in a scene. In this work, we tackle a fundamental issue of action binding in existing video diffusion models, which struggle to associate specific actions with their corresponding subjects. For this purpose, we propose ActionParty, an action controllable multi-subject world model for generative video games. It introduces subject state tokens, i.e. latent variables that persistently capture the state of each subject in the scene. By jointly modeling state tokens and video latents with a spatial biasing mechanism, we disentangle global video frame rendering from individual action-controlled subject updates. We evaluate ActionParty on the Melting Pot benchmark, demonstrating the first video world model capable of controlling up to seven players simultaneously across 46 diverse environments. Our results show significant improvements in action-following accuracy and identity consistency, while enabling robust autoregressive tracking of subjects through complex interactions. |
Proje...Project page: https://action-party.github.io/ |
| Omni123: Exploring 3D Native Foundation Models with Limited 3D Data by Unifying Text to 2D and 3D Generation | 2026-04-02 | ShowRecent multimodal large language models have achieved strong performance in unified text and image understanding and generation, yet extending such native capability to 3D remains challenging due to limited data. Compared to abundant 2D imagery, high-quality 3D assets are scarce, making 3D synthesis under-constrained. Existing methods often rely on indirect pipelines that edit in 2D and lift results into 3D via optimization, sacrificing geometric consistency. We present Omni123, a 3D-native foundation model that unifies text-to-2D and text-to-3D generation within a single autoregressive framework. Our key insight is that cross-modal consistency between images and 3D can serve as an implicit structural constraint. By representing text, images, and 3D as discrete tokens in a shared sequence space, the model leverages abundant 2D data as a geometric prior to improve 3D representations. We introduce an interleaved X-to-X training paradigm that coordinates diverse cross-modal tasks over heterogeneous paired datasets without requiring fully aligned text-image-3D triplets. By traversing semantic-visual-geometric cycles (e.g., text to image to 3D to image) within autoregressive sequences, the model jointly enforces semantic alignment, appearance fidelity, and multi-view geometric consistency. Experiments show that Omni123 significantly improves text-guided 3D generation and editing, demonstrating a scalable path toward multimodal 3D world models. |
|
| LatentUM: Unleashing the Potential of Interleaved Cross-Modal Reasoning via a Latent-Space Unified Model | 2026-04-02 | ShowUnified models (UMs) hold promise for their ability to understand and generate content across heterogeneous modalities. Compared to merely generating visual content, the use of UMs for interleaved cross-modal reasoning is more promising and valuable, e.g., for solving understanding problems that require dense visual thinking, improving visual generation through self-reflection, or modeling visual dynamics of the physical world guided by stepwise action interventions. However, existing UMs necessitate pixel decoding as a bridge due to their disjoint visual representations for understanding and generation, which is both ineffective and inefficient. In this paper, we introduce LatentUM, a novel unified model that represents all modalities within a shared semantic latent space, eliminating the need for pixel-space mediation between visual understanding and generation. This design naturally enables flexible interleaved cross-modal reasoning and generation. Beyond improved computational efficiency, the shared representation substantially alleviates codec bias and strengthens cross-modal alignment, allowing LatentUM to achieve state-of-the-art performance on the Visual Spatial Planning benchmark, push the limits of visual generation through self-reflection, and support world modeling by predicting future visual states within the shared semantic latent space. |
|
| World Action Verifier: Self-Improving World Models via Forward-Inverse Asymmetry | 2026-04-02 | ShowGeneral-purpose world models promise scalable policy evaluation, optimization, and planning, yet achieving the required level of robustness remains challenging. Unlike policy learning, which primarily focuses on optimal actions, a world model must be reliable over a much broader range of suboptimal actions, which are often insufficiently covered by action-labeled interaction data. To address this challenge, we propose World Action Verifier (WAV), a framework that enables world models to identify their own prediction errors and self-improve. The key idea is to decompose action-conditioned state prediction into two factors -- state plausibility and action reachability -- and verify each separately. We show that these verification problems can be substantially easier than predicting future states due to two underlying asymmetries: the broader availability of action-free data and the lower dimensionality of action-relevant features. Leveraging these asymmetries, we augment a world model with (i) a diverse subgoal generator obtained from video corpora and (ii) a sparse inverse model that infers actions from a subset of state features. By enforcing cycle consistency among generated subgoals, inferred actions, and forward rollouts, WAV provides an effective verification mechanism in under-explored regimes, where existing methods typically fail. Across nine tasks spanning MiniGrid, RoboMimic, and ManiSkill, our method achieves 2x higher sample efficiency while improving downstream policy performance by 18%. |
Proje...Project Website: https://world-action-verifier.github.io |
| DriveDreamer-Policy: A Geometry-Grounded World-Action Model for Unified Generation and Planning | 2026-04-02 | ShowRecently, world-action models (WAM) have emerged to bridge vision-language-action (VLA) models and world models, unifying their reasoning and instruction-following capabilities and spatio-temporal world modeling. However, existing WAM approaches often focus on modeling 2D appearance or latent representations, with limited geometric grounding-an essential element for embodied systems operating in the physical world. We present DriveDreamer-Policy, a unified driving world-action model that integrates depth generation, future video generation, and motion planning within a single modular architecture. The model employs a large language model to process language instructions, multi-view images, and actions, followed by three lightweight generators that produce depth, future video, and actions. By learning a geometry-aware world representation and using it to guide both future prediction and planning within a unified framework, the proposed model produces more coherent imagined futures and more informed driving actions, while maintaining modularity and controllable latency. Experiments on the Navsim v1 and v2 benchmarks demonstrate that DriveDreamer-Policy achieves strong performance on both closed-loop planning and world generation tasks. In particular, our model reaches 89.2 PDMS on Navsim v1 and 88.7 EPDMS on Navsim v2, outperforming existing world-model-based approaches while producing higher-quality future video and depth predictions. Ablation studies further show that explicit depth learning provides complementary benefits to video imagination and improves planning robustness. |
11 pa...11 pages, 4 figures; Project Website: https://drivedreamer-policy.github.io/ |
| ModTrans: Translating Real-world Models for Distributed Training Simulator | 2026-04-02 | ShowLarge-scale distributed training has been a research hot spot in machine learning systems for industry and academia in recent years. However, conducting experiments without physical machines and corresponding resources is difficult. One solution is to leverage distributed training simulators, but current ones like ASTRA-sim do not support importing real-world developed models, which poses challenges for ML researchers seeking to use them. Based on this challenge, we developed ModTrans, a translator supporting format translation from any real-world model to the ASTRA-sim simulator's input, removing the barrier between machine learning experts and machine learning system researchers. The experiment results show that ModTrans's cost is negligible. |
|
| F3DGS: Federated 3D Gaussian Splatting for Decentralized Multi-Agent World Modeling | 2026-04-02 | ShowWe present F3DGS, a federated 3D Gaussian Splatting framework for decentralized multi-agent 3D reconstruction. Existing 3DGS pipelines assume centralized access to all observations, which limits their applicability in distributed robotic settings where agents operate independently, and centralized data aggregation may be restricted. Directly extending centralized training to multi-agent systems introduces communication overhead and geometric inconsistency. F3DGS first constructs a shared geometric scaffold by registering locally merged LiDAR point clouds from multiple clients to initialize a global 3DGS model. During federated optimization, Gaussian positions are fixed to preserve geometric alignment, while each client updates only appearance-related attributes, including covariance, opacity, and spherical harmonic coefficients. The server aggregates these updates using visibility-aware aggregation, weighting each client's contribution by how frequently it observed each Gaussian, resolving the partial-observability challenge inherent to multi-agent exploration. To evaluate decentralized reconstruction, we collect a multi-sequence indoor dataset with synchronized LiDAR, RGB, and IMU measurements. Experiments show that F3DGS achieves reconstruction quality comparable to centralized training while enabling distributed optimization across agents. The dataset, development kit, and source code will be publicly released. |
Accep...Accepted to the CVPR 2026 SPAR-3D Workshop |
| What Capable Agents Must Know: Selection Theorems for Robust Decision-Making under Uncertainty | 2026-04-02 | ShowAs artificial agents become increasingly capable, what internal structure is necessary for an agent to act competently under uncertainty? Classical results show that optimal control can be implemented using belief states or world models, but not that such representations are required. We prove quantitative "selection theorems" showing that strong task performance (low average-case regret) forces world models, belief-like memory and -- under task mixtures -- persistent variables resembling core primitives associated with emotion, along with informational modularity under block-structured tasks. Our results cover stochastic policies, partial observability, and evaluation under task distributions, without assuming optimality, determinism, or access to an explicit model. Technically, we reduce predictive modeling to binary "betting" decisions and show that regret bounds limit probability mass on suboptimal bets, enforcing the predictive distinctions needed to separate high-margin outcomes. In fully observed settings, this yields approximate recovery of the interventional transition kernel; under partial observability, it implies necessity of predictive state and belief-like memory, addressing an open question in prior world-model recovery work. |
23 pa...23 pages; added PSR recovery (Theorems 3 & 4), and updated related work |
| Semantic Modeling for World-Centered Architectures | 2026-04-01 | ShowWe introduce world-centered multi-agent systems (WMAS) as an alternative to traditional agent-centered architectures, arguing that structured domains such as enterprises and institutional systems require a shared, explicit world representation to ensure semantic consistency, explainability, and long-term stability. We classify worlds along dimensions including ontological explicitness, normativity, etc. In WMAS, learning and coordination operate over a shared world model rather than isolated agent-local representations, enabling global consistency and verifiable system behavior. We propose semantic models as a mathematical formalism for representing such worlds. Finally, we present the Ontobox platform as a realization of WMAS. |
15 pa...15 pages, 1 figure, MathAI conference |
| Foundation Models for Autonomous Driving System: An Initial Roadmap | 2026-04-01 | ShowRecent advances in foundation models (FMs), including large language models (LLMs), vision-language models (VLMs), and world models, have opened new opportunities for autonomous driving systems (ADSs) in perception, reasoning, decision-making, and interaction. However, ADSs are safety-critical cyber-physical systems, and integrating FMs into them raises substantial software engineering challenges in data curation, system design, deployment, evaluation, and assurance. To clarify this rapidly evolving landscape, we present an initial roadmap, grounded in a structured literature review, for integrating FMs into autonomous driving across three dimensions: FM infrastructure, in-vehicle integration, and practical deployment. For each dimension, we summarize the state of the art, identify key challenges, and highlight open research opportunities. Based on this analysis, we outline research directions for building reliable, safe, and trustworthy FM-enabled ADSs. |
To ap...To appear in ACM Transactions on Software Engineering and Methodology (TOSEM) |
| Safety, Security, and Cognitive Risks in World Models | 2026-04-01 | ShowWorld models -- learned internal simulators of environment dynamics -- are rapidly becoming foundational to autonomous decision-making in robotics, autonomous vehicles, and agentic AI. Yet this predictive power introduces a distinctive set of safety, security, and cognitive risks. Adversaries can corrupt training data, poison latent representations, and exploit compounding rollout errors to cause catastrophic failures in safety-critical deployments. World model-equipped agents are more capable of goal misgeneralisation, deceptive alignment, and reward hacking precisely because they can simulate the consequences of their own actions. Authoritative world model predictions further foster automation bias and miscalibrated human trust that operators lack the tools to audit. This paper surveys the world model landscape; introduces formal definitions of trajectory persistence and representational risk; presents a five-profile attacker capability taxonomy; and develops a unified threat model extending MITRE ATLAS and the OWASP LLM Top 10 to the world model stack. We provide an empirical proof-of-concept on trajectory-persistent adversarial attacks (GRU-RSSM: A_1 = 2.26x amplification, -59.5% reduction under adversarial fine-tuning; stochastic RSSM proxy: A_1 = 0.65x; DreamerV3 checkpoint: non-zero action drift confirmed). We illustrate risks through four deployment scenarios and propose interdisciplinary mitigations spanning adversarial hardening, alignment engineering, NIST AI RMF and EU AI Act governance, and human-factors design. We argue that world models must be treated as safety-critical infrastructure requiring the same rigour as flight-control software or medical devices. |
26 pa...26 pages, 1 figure (6 panels), 2 tables. Empirical proof-of-concept on GRU/RSSM/DreamerV3 architectures |
| Deterministic World Models for Verification of Closed-loop Vision-based Systems | 2026-04-01 | ShowVerifying closed-loop vision-based control systems remains a fundamental challenge due to the high dimensionality of images and the difficulty of modeling visual environments. While generative models are increasingly used as camera surrogates in verification, their reliance on stochastic latent variables introduces unnecessary overapproximation error. To address this bottleneck, we propose a Deterministic World Model (DWM) that maps system states directly to generative images, effectively eliminating uninterpretable latent variables to ensure precise input bounds. The DWM is trained with a dual-objective loss function that combines pixel-level reconstruction accuracy with a control difference loss to maintain behavioral consistency with the real system. We integrate DWM into a verification pipeline utilizing Star-based reachability analysis (StarV) and employ conformal prediction to derive rigorous statistical bounds on the trajectory deviation between the world model and the actual vision-based system. Experiments on standard benchmarks show that our approach yields significantly tighter reachable sets and better verification performance than a latent-variable baseline. |
Signi...Significantly revised version with additional experiments and updated results. Submitted to EMSOFT 2026 |
| DLWM: Dual Latent World Models enable Holistic Gaussian-centric Pre-training in Autonomous Driving | 2026-04-01 | ShowVision-based autonomous driving has gained much attention due to its low costs and excellent performance. Compared with dense BEV (Bird's Eye View) or sparse query models, Gaussian-centric method is a comprehensive yet sparse representation by describing scene with 3D semantic Gaussians. In this paper, we introduce DLWM, a novel paradigm with Dual Latent World Models specifically designed to enable holistic gaussian-centric pre-training in autonomous driving using two stages. In the first stage, DLWM predicts 3D Gaussians from queries by self-supervised reconstructing multi-view semantic and depth images. Equipped with fine-grained contextual features, in the second stage, two latent world models are trained separately for temporal feature learning, including Gaussian-flow-guided latent prediction for downstream occupancy perception and forecasting tasks, and ego-planning-guided latent prediction for motion planning. Extensive experiments in SurroundOcc and nuScenes benchmarks demonstrate that DLWM shows significant performance gains across Gaussian-centric 3D occupancy perception, 4D occupancy forecasting and motion planning tasks. |
Accep...Accepted by CVPR 2026 |
| DreamerAD: Efficient Reinforcement Learning via Latent World Model for Autonomous Driving | 2026-04-01 | ShowWe introduce DreamerAD, the first latent world model framework that enables efficient reinforcement learning for autonomous driving by compressing diffusion sampling from 100 steps to 1 - achieving 80x speedup while maintaining visual interpretability. Training RL policies on real-world driving data incurs prohibitive costs and safety risks. While existing pixel-level diffusion world models enable safe imagination-based training, they suffer from multi-step diffusion inference latency (2s/frame) that prevents high-frequency RL interaction. Our approach leverages denoised latent features from video generation models through three key mechanisms: (1) shortcut forcing that reduces sampling complexity via recursive multi-resolution step compression, (2) an autoregressive dense reward model operating directly on latent representations for fine-grained credit assignment, and (3) Gaussian vocabulary sampling for GRPO that constrains exploration to physically plausible trajectories. DreamerAD achieves 87.7 EPDMS on NavSim v2, establishing state-of-the-art performance and demonstrating that latent-space RL is effective for autonomous driving. |
authors update |
| Toward Physically Consistent Driving Video World Models under Challenging Trajectories | 2026-04-01 | ShowVideo generation models have shown strong potential as world models for autonomous driving simulation. However, existing approaches are primarily trained on real-world driving datasets, which mostly contain natural and safe driving scenarios. As a result, current models often fail when conditioned on challenging or counterfactual trajectories-such as imperfect trajectories generated by simulators or planning systems-producing videos with severe physical inconsistencies and artifacts. To address this limitation, we propose PhyGenesis, a world model designed to generate driving videos with high visual fidelity and strong physical consistency. Our framework consists of two key components: (1) a physical condition generator that transforms potentially invalid trajectory inputs into physically plausible conditions, and (2) a physics-enhanced video generator that produces high-fidelity multi-view driving videos under these conditions. To effectively train these components, we construct a large-scale, physics-rich heterogeneous dataset. Specifically, in addition to real-world driving videos, we generate diverse challenging driving scenarios using the CARLA simulator, from which we derive supervision signals that guide the model to learn physically grounded dynamics under extreme conditions. This challenging-trajectory learning strategy enables trajectory correction and promotes physically consistent video generation. Extensive experiments demonstrate that PhyGenesis consistently outperforms state-of-the-art methods, especially on challenging trajectories. Our project page is available at: https://wm-research.github.io/PhyGenesis/. |
|
| Do World Action Models Generalize Better than VLAs? A Robustness Study | 2026-04-01 | ShowRobot action planning in the real world is challenging as it requires not only understanding the current state of the environment but also predicting how it will evolve in response to actions. Vision-language-action (VLA), which repurpose large-scale vision-language models for robot action generation using action experts, have achieved notable success across a variety of robotic tasks. Nevertheless, their performance remains constrained by the scope of their training data, exhibiting limited generalization to unseen scenarios and vulnerability to diverse contextual perturbations. More recently, world models have been revisited as an alternative to VLAs. These models, referred to as world action models (WAMs), are built upon world models that are trained on large corpora of video data to predict future states. With minor adaptations, their latent representation can be decoded into robot actions. It has been suggested that their explicit dynamic prediction capacity, combined with spatiotemporal priors acquired from web-scale video pretraining, enables WAMs to generalize more effectively than VLAs. In this paper, we conduct a comparative study of prominent state-of-the-art VLA policies and recently released WAMs. We evaluate their performance on the LIBERO-Plus and RoboTwin 2.0-Plus benchmarks under various visual and language perturbations. Our results show that WAMs achieve strong robustness, with LingBot-VA reaching 74.2% success rate on RoboTwin 2.0-Plus and Cosmos-Policy achieving 82.2% on LIBERO-Plus. While VLAs such as |
|
| Human-in-the-Loop Control of Objective Drift in LLM-Assisted Computer Science Education | 2026-03-31 | ShowLarge language models (LLMs) are increasingly embedded in computer science education through AI-assisted programming tools, yet such workflows often exhibit objective drift, in which locally plausible outputs diverge from stated task specifications. Existing instructional responses frequently emphasize tool-specific prompting practices, limiting durability as AI platforms evolve. This paper adopts a human-centered stance, treating human-in-the-loop (HITL) control as a stable educational problem rather than a transitional step toward AI autonomy. Drawing on systems engineering and control-theoretic concepts, we frame objectives and world models as operational artifacts that students configure to stabilize AI-assisted work. We propose a pilot undergraduate CS laboratory curriculum that explicitly separates planning from execution and trains students to specify acceptance criteria and architectural constraints prior to code generation. In selected labs, the curriculum also introduces deliberate, concept-aligned drift to support diagnosis and recovery from specification violations. We report a sensitivity power analysis for a three-arm pilot design comparing unstructured AI use, structured planning, and structured planning with injected drift, establishing detectable effect sizes under realistic section-level constraints. The contribution is a theory-driven, methodologically explicit foundation for HITL pedagogy that renders control competencies teachable across evolving AI tools. |
8 pages |
| SIMPACT: Simulation-Enabled Action Planning using Vision-Language Models | 2026-03-31 | ShowVision-Language Models (VLMs) exhibit remarkable common-sense and semantic reasoning capabilities. However, they lack a grounded understanding of physical dynamics. This limitation arises from training VLMs on static internet-scale visual-language data that contain no causal interactions or action-conditioned changes. Consequently, it remains challenging to leverage VLMs for fine-grained robotic manipulation tasks that require physical understanding, reasoning, and corresponding action planning. To overcome this, we present SIMPACT, a test-time, SIMulation-enabled ACTion Planning framework that equips VLMs with physical reasoning through simulation-in-the-loop world modeling, without requiring any additional training. From a single RGB-D observation, SIMPACT efficiently constructs physics simulations, enabling the VLM to propose informed actions, observe simulated rollouts, and iteratively refine its reasoning. By integrating language reasoning with physics prediction, our simulation-enabled VLM can understand contact dynamics and action outcomes in a physically grounded way. Our method demonstrates state-of-the-art performance on five challenging, real-world rigid-body and deformable manipulation tasks that require fine-grained physical reasoning, outperforming existing general-purpose robotic manipulation models. Our results demonstrate that embedding physics understanding via efficient simulation into VLM reasoning at test time offers a promising path towards generalizable embodied intelligence. Project webpage can be found at https://simpact-bot.github.io |
Accep...Accepted to CVPR 2026; camera-ready version |
| DIAL: Decoupling Intent and Action via Latent World Modeling for End-to-End VLA | 2026-03-31 | ShowThe development of Vision-Language-Action (VLA) models has been significantly accelerated by pre-trained Vision-Language Models (VLMs). However, most existing end-to-end VLAs treat the VLM primarily as a multimodal encoder, directly mapping vision-language features to low-level actions. This paradigm underutilizes the VLM's potential in high-level decision making and introduces training instability, frequently degrading its rich semantic representations. To address these limitations, we introduce DIAL, a framework bridging high-level decision making and low-level motor execution through a differentiable latent intent bottleneck. Specifically, a VLM-based System-2 performs latent world modeling by synthesizing latent visual foresight within the VLM's native feature space; this foresight explicitly encodes intent and serves as the structural bottleneck. A lightweight System-1 policy then decodes this predicted intent together with the current observation into precise robot actions via latent inverse dynamics. To ensure optimization stability, we employ a two-stage training paradigm: a decoupled warmup phase where System-2 learns to predict latent futures while System-1 learns motor control under ground-truth future guidance within a unified feature space, followed by seamless end-to-end joint optimization. This enables action-aware gradients to refine the VLM backbone in a controlled manner, preserving pre-trained knowledge. Extensive experiments on the RoboCasa GR1 Tabletop benchmark show that DIAL establishes a new state-of-the-art, achieving superior performance with 10x fewer demonstrations than prior methods. Furthermore, by leveraging heterogeneous human demonstrations, DIAL learns physically grounded manipulation priors and exhibits robust zero-shot generalization to unseen objects and novel configurations during real-world deployment on a humanoid robot. |
Proje...Project page: https://xpeng-robotics.github.io/dial |
| IndoorR2X: Indoor Robot-to-Everything Coordination with LLM-Driven Planning | 2026-03-31 | ShowAlthough robot-to-robot (R2R) communication improves indoor scene understanding beyond what a single robot can achieve, R2R alone cannot overcome partial observability without substantial exploration overhead or scaling team size. In contrast, many indoor environments already include low-cost Internet of Things (IoT) sensors (e.g., cameras) that provide persistent, building-wide context beyond onboard perception. We therefore introduce IndoorR2X, the first benchmark and simulation framework for Large Language Model (LLM)-driven multi-robot task planning with Robot-to-Everything (R2X) perception and communication in indoor environments. IndoorR2X integrates observations from mobile robots and static IoT devices to construct a global semantic state that supports scalable scene understanding, reduces redundant exploration, and enables high-level coordination through LLM-based planning. IndoorR2X provides configurable simulation environments, sensor layouts, robot teams, and task suites to systematically evaluate high-level semantic coordination strategies. Extensive experiments across diverse settings demonstrate that IoT-augmented world modeling improves multi-robot efficiency and reliability, and we highlight key insights and failure modes for advancing LLM-based collaboration between robot teams and indoor IoT sensors. See our project website: https://fandulu.github.io/IndoorR2X_project_page/. |
|
| SafeDrive: Fine-Grained Safety Reasoning for End-to-End Driving in a Sparse World | 2026-03-31 | ShowThe end-to-end (E2E) paradigm, which maps sensor inputs directly to driving decisions, has recently attracted significant attention due to its unified modeling capability and scalability. However, ensuring safety in this unified framework remains one of the most critical challenges. In this work, we propose SafeDrive, an E2E planning framework designed to perform explicit and interpretable safety reasoning through a trajectory-conditioned Sparse World Model. SafeDrive comprises two complementary networks: the Sparse World Network (SWNet) and the Fine-grained Reasoning Network (FRNet). SWNet constructs trajectory-conditioned sparse worlds that simulate the future behaviors of critical dynamic agents and road entities, providing interaction-centric representations for downstream reasoning. FRNet then evaluates agent-specific collision risks and temporal adherence to drivable regions, enabling precise identification of safety-critical events across future timesteps. SafeDrive achieves state-of-the-art performance on both open-loop and closed-loop benchmarks. On NAVSIM, it records a PDMS of 91.6 and an EPDMS of 87.5, with only 61 collisions out of 12,146 scenarios (0.5%). On Bench2Drive, SafeDrive attains a 66.8% driving score. |
Accep...Accepted to CVPR 2026, 19 pages, 9 figures |
| Towards High-Consistency Embodied World Model with Multi-View Trajectory Videos | 2026-03-31 | ShowEmbodied world models aim to predict and interact with the physical world through visual observations and actions. However, existing models struggle to accurately translate low-level actions (e.g., joint positions) into precise robotic movements in predicted frames, leading to inconsistencies with real-world physical interactions. To address these limitations, we propose MTV-World, an embodied world model that introduces Multi-view Trajectory-Video control for precise visuomotor prediction. Specifically, instead of directly using low-level actions for control, we employ trajectory videos obtained through camera intrinsic and extrinsic parameters and Cartesian-space transformation as control signals. However, projecting 3D raw actions onto 2D images inevitably causes a loss of spatial information, making a single view insufficient for accurate interaction modeling. To overcome this, we introduce a multi-view framework that compensates for spatial information loss and ensures high-consistency with physical world. MTV-World forecasts future frames based on multi-view trajectory videos as input and conditioning on an initial frame per view. Furthermore, to systematically evaluate both robotic motion precision and object interaction accuracy, we develop an auto-evaluation pipeline leveraging multimodal large models and referring video object segmentation models. To measure spatial consistency, we formulate it as an object location matching problem and adopt the Jaccard Index as the evaluation metric. Extensive experiments demonstrate that MTV-World achieves precise control execution and accurate physical interaction modeling in complex dual-arm scenarios. |
12 pages, 5 figures |
| The Mouth is Not the Brain: Bridging Energy-Based World Models and Language Generation | 2026-03-31 | ShowLarge Language Models (LLMs) generate fluent text, yet whether they truly understand the world or merely produce plausible texts about it remains contested. We propose an architectural principle, the mouth is not the brain, that explicitly separates world models from language models. Our architecture comprises three components: a DBM that captures domain structure as an energy-based world model, an adapter that projects latent belief states into embedding space, and a frozen GPT-2 that provides linguistic competence without domain knowledge. We instantiate this framework in the consumer review domain using Amazon smartphone reviews. Experiments demonstrate that (1) world model conditioning achieves lower cross-entropy loss and higher semantic similarity than architectural baselines including direct projection and full fine-tuning, while qualitative analysis reveals that soft prompt conditioning resolves a trade-off that prompt-based approaches cannot: simple prompts lack expressiveness while detailed prompts cause output collapse in small LLMs; (2) the DBM's energy function distinguishes coherent from incoherent market configurations, assigning higher energy to implausible brand-price combinations; and (3) interventions on specific attributes propagate causally to generated text with intervened outputs exhibiting distributions statistically consistent with naturally occurring samples sharing the target configuration. These findings suggest that even small-scale language models can achieve consistent, controllable generation when connected to an appropriate world model, providing empirical support for separating linguistic competence from world understanding. |
ICLR ...ICLR 2026 The 2nd Workshop on World Models: Understanding, Modelling, and Scaling |
| X-World: Controllable Ego-Centric Multi-Camera World Models for Scalable End-to-End Driving | 2026-03-31 | ShowScalable and reliable evaluation is increasingly critical in the end-to-end era of autonomous driving, where vision--language--action (VLA) policies directly map raw sensor streams to driving actions. Yet, current evaluation pipelines still rely heavily on real-world road testing, which is costly, biased toward limited scenario coverage, and difficult to reproduce. These challenges motivate a real-world simulator that can generate realistic future observations under proposed actions, while remaining controllable and stable over long horizons. We present X-World, an action-conditioned multi-camera generative world model that simulates future observations directly in video space. Given synchronized multi-view camera history and a future action sequence, X-World generates future multi-camera video streams that follow the commanded actions. To ensure reproducible and editable scene rollouts, X-World further supports optional controls over dynamic traffic agents and static road elements, and retains a text-prompt interface for appearance-level control (e.g., weather and time of day). Beyond world simulation, X-World also enables video style transfer by conditioning on appearance prompts while preserving the underlying action and scene dynamics. At the core of X-World is a multi-view latent video generator designed to explicitly encourage cross-view geometric consistency and temporal coherence under diverse control signals. Experiments show that X-World achieves high-quality multi-view video generation with (i) strong view consistency across cameras, (ii) stable temporal dynamics over long rollouts, and (iii) high controllability with strict action following and faithful adherence to optional scene controls. These properties make X-World a practical foundation for scalable and reproducible evaluation. |
Technical Report |
| HCLSM: Hierarchical Causal Latent State Machines for Object-Centric World Modeling | 2026-03-31 | ShowWorld models that predict future states from video remain limited by flat latent representations that entangle objects, ignore causal structure, and collapse temporal dynamics into a single scale. We present HCLSM, a world model architecture that operates on three interconnected principles: object-centric decomposition via slot attention with spatial broadcast decoding, hierarchical temporal dynamics through a three-level engine combining selective state space models for continuous physics, sparse transformers for discrete events, and compressed transformers for abstract goals, and causal structure learning through graph neural network interaction patterns. HCLSM introduces a two-stage training protocol where spatial reconstruction forces slot specialization before dynamics prediction begins. We train a 68M-parameter model on the PushT robotic manipulation benchmark from the Open X-Embodiment dataset, achieving 0.008 MSE next-state prediction loss with emerging spatial decomposition (SBD loss: 0.0075) and learned event boundaries. A custom Triton kernel for the SSM scan delivers 38x speedup over sequential PyTorch. The full system spans 8,478 lines of Python across 51 modules with 171 unit tests. Code: https://github.com/rightnow-ai/hclsm |
10 pa...10 pages, 3 tables, 4 figures, 1 algorithm. Code: https://github.com/rightnow-ai/hclsm |
| MultiGen: Level-Design for Editable Multiplayer Worlds in Diffusion Game Engines | 2026-03-30 | ShowVideo world models have shown immense promise for interactive simulation and entertainment, but current systems still struggle with two important aspects of interactivity: user control over the environment for reproducible, editable experiences, and shared inference where players hold influence over a common world. To address these limitations, we introduce an explicit external memory into the system, a persistent state operating independent of the model's context window, that is continually updated by user actions and queried throughout the generation roll-out. Unlike conventional diffusion game engines that operate as next-frame predictors, our approach decomposes generation into Memory, Observation, and Dynamics modules. This design gives users direct, editable control over environment structure via an editable memory representation, and it naturally extends to real-time multiplayer rollouts with coherent viewpoints and consistent cross-player interactions. |
Proje...Project page here: https://ryanpo.com/multigen/ |
| Stepper: Stepwise Immersive Scene Generation with Multiview Panoramas | 2026-03-30 | ShowThe synthesis of immersive 3D scenes from text is rapidly maturing, driven by novel video generative models and feed-forward 3D reconstruction, with vast potential in AR/VR and world modeling. While panoramic images have proven effective for scene initialization, existing approaches suffer from a trade-off between visual fidelity and explorability: autoregressive expansion suffers from context drift, while panoramic video generation is limited to low resolution. We present Stepper, a unified framework for text-driven immersive 3D scene synthesis that circumvents these limitations via stepwise panoramic scene expansion. Stepper leverages a novel multi-view 360° diffusion model that enables consistent, high-resolution expansion, coupled with a geometry reconstruction pipeline that enforces geometric coherence. Trained on a new large-scale, multi-view panorama dataset, Stepper achieves state-of-the-art fidelity and structural consistency, outperforming prior approaches, thereby setting a new standard for immersive scene generation. |
Accep...Accepted at CVPR 2026 Findings; Find our project page under https://fwmb.github.io/stepper/ |
| AutoWorld: Scaling Multi-Agent Traffic Simulation with Self-Supervised World Models | 2026-03-30 | ShowMulti-agent traffic simulation is central to developing and testing autonomous driving systems. Recent data-driven simulators have achieved promising results, but rely heavily on supervised learning from labeled trajectories or semantic annotations, making it costly to scale their performance. Meanwhile, large amounts of unlabeled sensor data can be collected at scale but remain largely unused by existing traffic simulation frameworks. This raises a key question: How can a method harness unlabeled data to improve traffic simulation performance? In this work, we propose AutoWorld, a traffic simulation framework that employs a world model learned from unlabeled occupancy representations of LiDAR data. Given world model samples, AutoWorld constructs a coarse-to-fine predictive scene context as input to a multi-agent motion generation model. To promote sample diversity, AutoWorld uses a cascaded Determinantal Point Process framework to guide the sampling processes of both the world model and the motion model. Furthermore, we designed a motion-aware latent supervision objective that enhances AutoWorld's representation of scene dynamics. Experiments on the WOSAC benchmark show that AutoWorld ranks first on the leaderboard according to the primary Realism Meta Metric (RMM). We further show that simulation performance consistently improves with the inclusion of unlabeled LiDAR data, and study the efficacy of each component with ablations. Our method paves the way for scaling traffic simulation realism without additional labeling. Our project page contains additional visualizations and released code. |
|
| Enhancing Policy Learning with World-Action Model | 2026-03-30 | ShowThis paper presents the World-Action Model (WAM), an action-regularized world model that jointly reasons over future visual observations and the actions that drive state transitions. Unlike conventional world models trained solely via image prediction, WAM incorporates an inverse dynamics objective into DreamerV2 that predicts actions from latent state transitions, encouraging the learned representations to capture action-relevant structure critical for downstream control. We evaluate WAM on enhancing policy learning across eight manipulation tasks from the CALVIN benchmark. We first pretrain a diffusion policy via behavioral cloning on world model latents, then refine it with model-based PPO inside the frozen world model. Without modifying the policy architecture or training procedure, WAM improves average behavioral cloning success from 59.4% to 71.2% over DreamerV2 and DiWA baselines. After PPO fine-tuning, WAM achieves 92.8% average success versus 79.8% for the baseline, with two tasks reaching 100%, using 8.7x fewer training steps. |
|
| OccSim: Multi-kilometer Simulation with Long-horizon Occupancy World Models | 2026-03-30 | ShowData-driven autonomous driving simulation has long been constrained by its heavy reliance on pre-recorded driving logs or spatial priors, such as HD maps. This fundamental dependency severely limits scalability, restricting open-ended generation capabilities to the finite scale of existing collected datasets. To break this bottleneck, we present OccSim, the first occupancy world model-driven 3D simulator. OccSim obviates the requirement for continuous logs or HD maps; conditioned only on a single initial frame and a sequence of future ego-actions, it can stably generate over 3,000 continuous frames, enabling the continuous construction of large-scale 3D occupancy maps spanning over 4 kilometers for simulation. This represents an >80x improvement in stable generation length over previous state-of-the-art occupancy world models. OccSim is powered by two modules: W-DiT based static occupancy world model and the Layout Generator. W-DiT handles the ultra-long-horizon generation of static environments by explicitly introducing known rigid transformations in architecture design, while the Layout Generator populates the dynamic foreground with reactive agents based on the synthesized road topology. With these designs, OccSim can synthesize massive, diverse simulation streams. Extensive experiments demonstrate its downstream utility: data collected directly from OccSim can pre-train 4D semantic occupancy forecasting models to achieve up to 67% zero-shot performance on unseen data, outperforming previous asset-based simulator by 11%. When scaling the OccSim dataset to 5x the size, the zero-shot performance increases to about 74%, while the improvement over asset-based simulators expands to 22.1%. |
|
| ManipArena: Comprehensive Real-world Evaluation of Reasoning-Oriented Generalist Robot Manipulation | 2026-03-30 | ShowVision-Language-Action (VLA) models and world models have recently emerged as promising paradigms for general-purpose robotic intelligence, yet their progress is hindered by the lack of reliable evaluation protocols that reflect real-world deployment. Existing benchmarks are largely simulator-centric, which provide controllability but fail to capture the reality gap caused by perception noise, complex contact dynamics, hardware constraints, and system latency. Moreover, fragmented real-world evaluations across different robot platforms prevent fair and reproducible comparison. To address these challenges, we introduce ManipArena, a standardized evaluation framework designed to bridge simulation and real-world execution. ManipArena comprises 20 diverse tasks across 10,812 expert trajectories emphasizing reasoning-oriented manipulation tasks requiring semantic and spatial reasoning, supports multi-level generalization through controlled out-of-distribution settings, and incorporates long-horizon mobile manipulation beyond tabletop scenarios. The framework further provides rich sensory diagnostics, including low-level motor signals, and synchronized real-to-sim environments constructed via high-quality 3D scanning. Together, these features enable fair, realistic, and reproducible evaluation for both VLA and world model approaches, providing a scalable foundation for diagnosing and advancing embodied intelligence systems. |
Techn...Technical report for CVPR 2026 Challenge ManipArena |
| Vega: Learning to Drive with Natural Language Instructions | 2026-03-30 | ShowVision-language-action models have reshaped autonomous driving to incorporate languages into the decision-making process. However, most existing pipelines only utilize the language modality for scene descriptions or reasoning and lack the flexibility to follow diverse user instructions for personalized driving. To address this, we first construct a large-scale driving dataset (InstructScene) containing around 100,000 scenes annotated with diverse driving instructions with the corresponding trajectories. We then propose a unified Vision-Language-World-Action model, Vega, for instruction-based generation and planning. We employ the autoregressive paradigm to process visual inputs (vision) and language instructions (language) and the diffusion paradigm to generate future predictions (world modeling) and trajectories (action). We perform joint attention to enable interactions between the modalities and use individual projection layers for different modalities for more capabilities. Extensive experiments demonstrate that our method not only achieves superior planning performance but also exhibits strong instruction-following abilities, paving the way for more intelligent and personalized driving systems. |
Code ...Code is available at https://github.com/zuosc19/Vega |
| Dream to Recall: Imagination-Guided Experience Retrieval for Memory-Persistent Vision-and-Language Navigation | 2026-03-30 | ShowVision-and-Language Navigation (VLN) requires agents to follow natural language instructions through environments, with memory-persistent variants demanding progressive improvement through accumulated experience. Existing approaches for memory-persistent VLN face critical limitations: they lack effective memory access mechanisms, instead relying on entire memory incorporation or fixed-horizon lookup, and predominantly store only environmental observations while neglecting navigation behavioral patterns that encode valuable decision-making strategies. We present Memoir, which employs imagination as a retrieval mechanism grounded by explicit memory: a world model imagines future navigation states as queries to selectively retrieve relevant environmental observations and behavioral histories. The approach comprises: 1) a language-conditioned world model that imagines future states serving dual purposes: encoding experiences for storage and generating retrieval queries; 2) Hybrid Viewpoint-Level Memory that anchors both observations and behavioral patterns to viewpoints, enabling hybrid retrieval; and 3) an experience-augmented navigation model that integrates retrieved knowledge through specialized encoders. Extensive evaluation across diverse memory-persistent VLN benchmarks with 10 distinct testing scenarios demonstrates Memoir's effectiveness: significant improvements across all scenarios, with 5.4% SPL gains on IR2R over the best memory-persistent baseline, accompanied by 8.3x training speedup and 74% inference memory reduction. The results validate that predictive retrieval of both environmental and behavioral memories enables more effective navigation, with analysis indicating substantial headroom (73.3% vs 93.4% upper bound) for this imagination-guided paradigm. |
Accep...Accepted by IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI) |
| Object-Centric World Models for Causality-Aware Reinforcement Learning | 2026-03-30 | ShowWorld models have been developed to support sample-efficient deep reinforcement learning agents. However, it remains challenging for world models to accurately replicate environments that are high-dimensional, non-stationary, and composed of multiple objects with rich interactions since most world models learn holistic representations of all environmental components. By contrast, humans perceive the environment by decomposing it into discrete objects, facilitating efficient decision-making. Motivated by this insight, we propose \emph{Slot Transformer Imagination with CAusality-aware reinforcement learning} (STICA), a unified framework in which object-centric Transformers serve as the world model and causality-aware policy and value networks. STICA represents each observation as a set of object-centric tokens, together with tokens for the agent action and the resulting reward, enabling the world model to predict token-level dynamics and interactions. The policy and value networks then estimate token-level cause--effect relations and use them in the attention layers, yielding causality-guided decision-making. Experiments on object-rich benchmarks demonstrate that STICA consistently outperforms state-of-the-art agents in both sample efficiency and final performance. |
Accep...Accepted by AAAI-26. Codes are available at https://github.com/nishimoto0430/STICA |
| InternVideo-Next: Towards General Video Foundation Models without Video-Text Supervision | 2026-03-30 | ShowLarge-scale video-text pretraining achieves strong performance but depends on noisy, synthetic captions with limited semantic coverage, often overlooking implicit world knowledge such as object motion, 3D geometry, and physical cues. In contrast, masked video modeling (MVM) directly exploits spatiotemporal structures but trails text-supervised methods on general tasks. We find this gap arises from overlooked architectural issues: pixel-level reconstruction struggles with convergence and its low-level requirement often conflicts with semantics, while latent prediction often encourages shortcut learning. To address these, we disentangle the traditional encoder-decoder design into an Encoder-Predictor-Decoder (EPD) framework, where the predictor acts as a latent world model, and propose InternVideo-Next, a two-stage pretraining scheme that builds a semantically consistent yet detail-preserving latent space for this world model. First, conventional linear decoder in pixel MVM enforces the predictor output latent to be linearly projected to, thus separable in pixel space, causing the conflict with semantic abstraction. Our Stage 1 proposes a conditional diffusion decoder and injects reliable image-level semantic priors to enhance semantics and convergence, thus bridging pixel-level fidelity with high-level semantic abstraction. Stage 2 further learns world knowledge by predicting frozen Stage 1 targets within this space, mitigating shortcut learning. Trained on public, unlabeled videos, InternVideo-Next achieves state-of-the-art results across benchmarks and provides a scalable path toward general video representation learning. |
|
| VerseCrafter: Dynamic Realistic Video World Model with 4D Geometric Control | 2026-03-30 | ShowVideo world models aim to simulate dynamic, real-world environments, yet existing methods struggle to provide unified and precise control over camera and multi-object motion, as videos inherently capture dynamics in the projected 2D image plane. To bridge this gap, we introduce VerseCrafter, a geometry-driven video world model that generates dynamic, realistic videos from a unified 4D geometric world state. Our approach is centered on a novel 4D Geometric Control representation, which encodes the world state as a static background point cloud and per-object 3D Gaussian trajectories. This representation captures each object's motion path and probabilistic 3D occupancy over time, providing a flexible, category-agnostic alternative to rigid bounding boxes and parametric models. We render 4D Geometric Control into 4D control maps for a pretrained video diffusion model, enabling high-fidelity, view-consistent video generation that faithfully follows the specified dynamics. To enable training at scale, we develop an automatic data engine and construct VerseControl4D, a real-world dataset of 35K training samples with automatically derived prompts and rendered 4D control maps. Extensive experiments show that VerseCrafter achieves superior visual quality and more accurate control over camera and multi-object motion than prior methods. |
Proje...Project Page: https://sixiaozheng.github.io/VerseCrafter_page/, Accepted by CVPR 2026 |
| Goal-VLA: Image-Generative VLMs as Object-Centric World Models Empowering Zero-shot Robot Manipulation | 2026-03-30 | ShowGeneralization remains a fundamental challenge in robotic manipulation. To tackle this challenge, recent Vision-Language-Action (VLA) models build policies on top of Vision-Language Models (VLMs), seeking to transfer their open-world semantic knowledge. However, their zero-shot capability lags significantly behind the base VLMs, as the instruction-vision-action data is too limited to cover diverse scenarios, tasks, and robot embodiments. In this work, we present Goal-VLA, a zero-shot framework that leverages Image-Generative VLMs as world models to generate desired goal states, from which the target object pose is derived to enable generalizable manipulation. The key insight is that object state representation is the golden interface, naturally separating a manipulation system into high-level and low-level policies. This representation abstracts away explicit action annotations, allowing the use of highly generalizable VLMs while simultaneously providing spatial cues for training-free low-level control. To further improve robustness, we introduce a Reflection-through-Synthesis process that iteratively validates and refines the generated goal image before execution. Both simulated and real-world experiments demonstrate that our \name achieves strong performance and inspiring generalizability in manipulation tasks. Supplementary materials are available at https://nus-lins-lab.github.io/goalvlaweb/. |
|
| ProgressVLA: Progress-Guided Diffusion Policy for Vision-Language Robotic Manipulation | 2026-03-29 | ShowMost existing vision-language-action (VLA) models for robotic manipulation lack progress awareness, typically relying on hand-crafted heuristics for task termination. This limitation is particularly severe in long-horizon tasks involving cascaded sub-goals. In this work, we investigate the estimation and integration of task progress, proposing a novel model named {\textbf \vla}. Our technical contributions are twofold: (1) \emph{robust progress estimation}: We pre-train a progress estimator on large-scale, unsupervised video-text robotic datasets. This estimator achieves a low prediction residual (0.07 on a scale of |
|
| LOME: Learning Human-Object Manipulation with Action-Conditioned Egocentric World Model | 2026-03-28 | ShowLearning human-object manipulation presents significant challenges due to its fine-grained and contact-rich nature of the motions involved. Traditional physics-based animation requires extensive modeling and manual setup, and more importantly, it neither generalizes well across diverse object morphologies nor scales effectively to real-world environment. To address these limitations, we introduce LOME, an egocentric world model that can generate realistic human-object interactions as videos conditioned on an input image, a text prompt, and per-frame human actions, including both body poses and hand gestures. LOME injects strong and precise action guidance into object manipulation by jointly estimating spatial human actions and the environment contexts during training. After finetuning a pretrained video generative model on videos of diverse egocentric human-object interactions, LOME demonstrates not only high action-following accuracy and strong generalization to unseen scenarios, but also realistic physical consequences of hand-object interactions, e.g., liquid flowing from a bottle into a mug after executing a ``pouring'' action. Extensive experiments demonstrate that our video-based framework significantly outperforms state-of-the-art image based and video-based action-conditioned methods and Image/Text-to-Video (I/T2V) generative model in terms of both temporal consistency and motion control. LOME paves the way for photorealistic AR/VR experiences and scalable robotic training, without being limited to simulated environments or relying on explicit 3D/4D modeling. |
|
| VLM-SAFE: Vision-Language Model-Guided Safety-Aware Reinforcement Learning with World Models for Autonomous Driving | 2026-03-28 | ShowAutonomous driving policy learning with reinforcement learning (RL) is fundamentally limited by low sample efficiency, weak generalization, and a dependence on unsafe online trial-and-error interactions. Although safe RL introduces explicit constraints or costs, existing methods often fail to capture the semantic meaning of safety in real driving scenes, leading to conservative behaviors in simple cases and insufficient risk awareness in complex ones. To address this issue, we propose VLM-SAFE, an offline safe RL framework that follows a human cognitive loop of observe-imagine-evaluate-act. Starting from offline driving data, VLM-SAFE observes traffic scenarios and leverages a vision-language model (VLM) to provide semantic safety signals grounded in scene understanding. A learned world model then imagines future trajectories from the observed context, enabling the agent to reason about possible consequences without interacting with the real environment. Rather than using imagined rollouts solely for return estimation, VLM-SAFE further evaluates these predicted futures with VLM-based safety guidance, explicitly coupling future anticipation with semantic risk assessment. The resulting safety-aware imagined experience is finally used to optimize the policy via actor-critic learning, such that actions are chosen based on both predicted outcomes and their safety implications. By tightly integrating observation, imagination, evaluation, and action into a unified closed loop, VLM-SAFE enables safer and more efficient offline policy learning for autonomous driving. Extensive experiments in simulation show that VLM-SAFE achieves improved safety, stronger robustness under traffic-density shift, and a better safety-performance trade-off than representative baselines. |
N/A |
| Uni-World VLA: Interleaved World Modeling and Planning for Autonomous Driving | 2026-03-28 | ShowAutonomous driving requires reasoning about how the environment evolves and planning actions accordingly. Existing world-model-based approaches typically predict future scenes first and plan afterwards, resulting in open-loop imagination that may drift from the actual decision process. In this paper, we present Uni-World VLA, a unified vision-language-action (VLA) model that tightly interleaves future frame prediction and trajectory planning. Instead of generating a full world rollout before planning, our model alternates between predicting future frames and ego actions step by step, allowing planning decisions to be continuously conditioned on the imagined future observations. This interleaved generation forms a closed-loop interaction between world modeling and control, enabling more adaptive decision-making in dynamic traffic scenarios. In addition, we incorporate monocular depth information into frames to provide stronger geometric cues for world modeling, improving long-horizon scene prediction. Experiments on the NAVSIM benchmark show that our approach achieves competitive closed-loop planning performance while producing high-fidelity future frame predictions. These results demonstrate that tightly coupling world prediction and planning is a promising direction for scalable VLA driving systems. |
22 pa...22 pages, 8 figures. Submitted to ECCV 2026. Code will be released |
| LLMs versus the Halting Problem: Revisiting Program Termination Prediction | 2026-03-28 | ShowDetermining whether a program terminates is a central problem in computer science. Turing's foundational result established the Halting Problem as undecidable, showing that no algorithm can universally determine termination for all programs and inputs. Consequently, automatic verification tools approximate termination, sometimes failing to prove or disprove; these tools rely on problem-specific architectures, and are usually tied to particular programming languages. Recent success and progress in large language models (LLMs) raises the following question: can LLMs reliably predict program termination? In this work, we evaluate LLMs on a diverse set of programs from the Termination category of the International Competition on Software Verification (SV-Comp) 2025. Our results suggest that LLMs perform remarkably well at predicting program termination, where GPT-5 and Claude Sonnet-4.5 would rank just behind the top-ranked tool (using test-time-scaling), and Code World Model (CWM) would place just behind the second-ranked tool. While LLMs are effective at predicting program termination, they often fail to provide a valid witness as a proof. Moreover, LLMs performance drops as program length and complexity increases. We hope these insights motivate further research into program termination and the broader potential of LLMs for reasoning about undecidable problems. |
|
| Out of Sight but Not Out of Mind: Hybrid Memory for Dynamic Video World Models | 2026-03-28 | ShowVideo world models have shown immense potential in simulating the physical world, yet existing memory mechanisms primarily treat environments as static canvases. When dynamic subjects hide out of sight and later re-emerge, current methods often struggle, leading to frozen, distorted, or vanishing subjects. To address this, we introduce Hybrid Memory, a novel paradigm requiring models to simultaneously act as precise archivists for static backgrounds and vigilant trackers for dynamic subjects, ensuring motion continuity during out-of-view intervals. To facilitate research in this direction, we construct HM-World, the first large-scale video dataset dedicated to hybrid memory. It features 59K high-fidelity clips with decoupled camera and subject trajectories, encompassing 17 diverse scenes, 49 distinct subjects, and meticulously designed exit-entry events to rigorously evaluate hybrid coherence. Furthermore, we propose HyDRA, a specialized memory architecture that compresses memory into tokens and utilizes a spatiotemporal relevance-driven retrieval mechanism. By selectively attending to relevant motion cues, HyDRA effectively preserves the identity and motion of hidden subjects. Extensive experiments on HM-World demonstrate that our method significantly outperforms state-of-the-art approaches in both dynamic subject consistency and overall generation quality. Code is publicly available at https://github.com/H-EmbodVis/HyDRA. |
Proje...Project Page: https://kj-chen666.github.io/Hybrid-Memory-in-Video-World-Models/ Code: https://github.com/H-EmbodVis/HyDRA |
| MMaDA-VLA: Large Diffusion Vision-Language-Action Model with Unified Multi-Modal Instruction and Generation | 2026-03-27 | ShowVision-Language-Action (VLA) models aim to control robots for manipulation from visual observations and natural-language instructions. However, existing hierarchical and autoregressive paradigms often introduce architectural overhead, suffer from temporal inconsistency and long-horizon error accumulation, and lack a mechanism to capture environment dynamics without extra modules. To this end, we present MMaDA-VLA, a fully native pre-trained large diffusion VLA model that unifies multi-modal understanding and generation in a single framework. Our key idea is a native discrete diffusion formulation that embeds language, images, and continuous robot controls into one discrete token space and trains a single backbone with masked token denoising to jointly generate a future goal observation and an action chunk in parallel. Iterative denoising enables global, order-free refinement, improving long-horizon consistency while grounding actions in predicted future visual outcomes without auxiliary world models. Experiments across simulation benchmarks and real-world tasks show state-of-the-art performance, achieving 98.0% average success on LIBERO and 4.78 average length on CALVIN. |
|
| CLARITY: Medical World Model for Guiding Treatment Decisions by Modeling Context-Aware Disease Trajectories in Latent Space | 2026-03-27 | ShowClinical decision-making in oncology requires predicting dynamic disease evolution, a task current static AI predictors cannot perform. While world models (WMs) offer a paradigm for generative prediction, existing medical applications remain limited. Existing methods often rely on stochastic diffusion models, focusing on visual reconstruction rather than causal, physiological transitions. Furthermore, in medical domain, models like MeWM typically ignore patient-specific temporal and clinical contexts and lack a feedback mechanism to link predictions to treatment decisions. To address these gaps, we introduce CLARITY, a medical world model that forecasts disease evolution directly within a structured latent space. It explicitly integrates time intervals (temporal context) and patient-specific data (clinical context) to model treatment-conditioned progression as a smooth, interpretable trajectory, and thus generate physiologically faithful, individualized treatment plans. Finally, CLARITY introduces a novel prediction-to-decision framework, translating latent rollouts into transparent, actionable recommendations. CLARITY demonstrates state-of-the-art performance in treatment planning. On the MU-Glioma-Post dataset, our approach outperforms recent MeWM by 12%, and significantly surpasses all other medical-specific large language models. |
|
| ABot-PhysWorld: Interactive World Foundation Model for Robotic Manipulation with Physics Alignment | 2026-03-27 | ShowVideo-based world models offer a powerful paradigm for embodied simulation and planning, yet state-of-the-art models often generate physically implausible manipulations - such as object penetration and anti-gravity motion - due to training on generic visual data and likelihood-based objectives that ignore physical laws. We present ABot-PhysWorld, a 14B Diffusion Transformer model that generates visually realistic, physically plausible, and action-controllable videos. Built on a curated dataset of three million manipulation clips with physics-aware annotation, it uses a novel DPO-based post-training framework with decoupled discriminators to suppress unphysical behaviors while preserving visual quality. A parallel context block enables precise spatial action injection for cross-embodiment control. To better evaluate generalization, we introduce EZSbench, the first training-independent embodied zero-shot benchmark combining real and synthetic unseen robot-task-scene combinations. It employs a decoupled protocol to separately assess physical realism and action alignment. ABot-PhysWorld achieves new state-of-the-art performance on PBench and EZSbench, surpassing Veo 3.1 and Sora v2 Pro in physical plausibility and trajectory consistency. We will release EZSbench to promote standardized evaluation in embodied video generation. |
Code:... |
| Any4D: Open-Prompt 4D Generation from Natural Language and Images | 2026-03-27 | ShowWhile video-generation-based embodied world models have gained increasing attention, their reliance on large-scale embodied interaction data remains a key bottleneck. The scarcity, difficulty of collection, and high dimensionality of embodied data fundamentally limit the alignment granularity between language and actions and exacerbate the challenge of long-horizon video generation--hindering generative models from achieving a \textit{"GPT moment"} in the embodied domain. There is a naive observation: \textit{the diversity of embodied data far exceeds the relatively small space of possible primitive motions}. Based on this insight, we propose \textbf{Primitive Embodied World Models} (PEWM), which restricts video generation to fixed shorter horizons, our approach \textit{1) enables} fine-grained alignment between linguistic concepts and visual representations of robotic actions, \textit{2) reduces} learning complexity, \textit{3) improves} data efficiency in embodied data collection, and \textit{4) decreases} inference latency. By equipping with a modular Vision-Language Model (VLM) planner and a Start-Goal heatmap Guidance mechanism (SGG), PEWM further enables flexible closed-loop control and supports compositional generalization of primitive-level policies over extended, complex tasks. Our framework leverages the spatiotemporal vision priors in video models and the semantic awareness of VLMs to bridge the gap between fine-grained physical interaction and high-level reasoning, paving the way toward scalable, interpretable, and general-purpose embodied intelligence. |
The a...The authors identified issues in the 4D generation pipeline and evaluation that affect result validity. To ensure scientific accuracy, we will revise the methodology and experiments thoroughly before resubmitting. This version should not be cited or relied upon |
| The Pulse of Motion: Measuring Physical Frame Rate from Visual Dynamics | 2026-03-27 | ShowWhile recent generative video models have achieved remarkable visual realism and are being explored as world models, true physical simulation requires mastering both space and time. Current models can produce visually smooth kinematics, yet they lack a reliable internal motion pulse to ground these motions in a consistent, real-world time scale. This temporal ambiguity stems from the common practice of indiscriminately training on videos with vastly different real-world speeds, forcing them into standardized frame rates. This leads to what we term chronometric hallucination: generated sequences exhibit ambiguous, unstable, and uncontrollable physical motion speeds. To address this, we propose Visual Chronometer, a predictor that recovers the Physical Frames Per Second (PhyFPS) directly from the visual dynamics of an input video. Trained via controlled temporal resampling, our method estimates the true temporal scale implied by the motion itself, bypassing unreliable metadata. To systematically quantify this issue, we establish two benchmarks, PhyFPS-Bench-Real and PhyFPS-Bench-Gen. Our evaluations reveal a harsh reality: state-of-the-art video generators suffer from severe PhyFPS misalignment and temporal instability. Finally, we demonstrate that applying PhyFPS corrections significantly improves the human-perceived naturalness of AI-generated videos. Our project page is https://xiangbogaobarry.github.io/Visual_Chronometer/. |
|
| Policy-Guided World Model Planning for Language-Conditioned Visual Navigation | 2026-03-26 | ShowNavigating to a visually specified goal given natural language instructions remains a fundamental challenge in embodied AI. Existing approaches either rely on reactive policies that struggle with long-horizon planning, or employ world models that suffer from poor action initialization in high-dimensional spaces. We present PiJEPA, a two-stage framework that combines the strengths of learned navigation policies with latent world model planning for instruction-conditioned visual navigation. In the first stage, we finetune an Octo-based generalist policy, augmented with a frozen pretrained vision encoder (DINOv2 or V-JEPA-2), on the CAST navigation dataset to produce an informed action distribution conditioned on the current observation and language instruction. In the second stage, we use this policy-derived distribution to warm-start Model Predictive Path Integral (MPPI) planning over a separately trained JEPA world model, which predicts future latent states in the embedding space of the same frozen encoder. By initializing the MPPI sampling distribution from the policy prior rather than from an uninformed Gaussian, our planner converges faster to high-quality action sequences that reach the goal. We systematically study the effect of the vision encoder backbone, comparing DINOv2 and V-JEPA-2, across both the policy and world model components. Experiments on real-world navigation tasks demonstrate that PiJEPA significantly outperforms both standalone policy execution and uninformed world model planning, achieving improved goal-reaching accuracy and instruction-following fidelity. |
|
| Out-of-Sight Embodied Agents: Multimodal Tracking, Sensor Fusion, and Trajectory Forecasting | 2026-03-26 | ShowTrajectory prediction is a fundamental problem in computer vision, vision-language-action models, world models, and autonomous systems, with broad impact on autonomous driving, robotics, and surveillance. However, most existing methods assume complete and clean observations, and therefore do not adequately handle out-of-sight agents or noisy sensing signals caused by limited camera coverage, occlusions, and the absence of ground-truth denoised trajectories. These challenges raise safety concerns and reduce robustness in real-world deployment. In this extended study, we introduce major improvements to Out-of-Sight Trajectory (OST), a task for predicting noise-free visual trajectories of out-of-sight objects from noisy sensor observations. Building on our prior work, we expand Out-of-Sight Trajectory Prediction (OOSTraj) from pedestrians to both pedestrians and vehicles, increasing its relevance to autonomous driving, robotics, and surveillance. Our improved Vision-Positioning Denoising Module exploits camera calibration to establish vision-position correspondence, mitigating the lack of direct visual cues and enabling effective unsupervised denoising of noisy sensor signals. Extensive experiments on the Vi-Fi and JRDB datasets show that our method achieves state-of-the-art results for both trajectory denoising and trajectory prediction, with clear gains over prior baselines. We also compare with classical denoising methods, including Kalman filtering, and adapt recent trajectory prediction models to this setting, establishing a stronger benchmark. To the best of our knowledge, this is the first work to use vision-positioning projection to denoise noisy sensor trajectories of out-of-sight agents, opening new directions for future research. |
Publi...Published in IEEE Transactions on Pattern Analysis and Machine Intelligence (Early Access), pp. 1-14, March 23, 2026 |
| World Reasoning Arena | 2026-03-26 | ShowWorld models (WMs) are intended to serve as internal simulators of the real world that enable agents to understand, anticipate, and act upon complex environments. Existing WM benchmarks remain narrowly focused on next-state prediction and visual fidelity, overlooking the richer simulation capabilities required for intelligent behavior. To address this gap, we introduce WR-Arena, a comprehensive benchmark for evaluating WMs along three fundamental dimensions of next world simulation: (i) Action Simulation Fidelity, the ability to interpret and follow semantically meaningful, multi-step instructions and generate diverse counterfactual rollouts; (ii) Long-horizon Forecast, the ability to sustain accurate, coherent, and physically plausible simulations across extended interactions; and (iii) Simulative Reasoning and Planning, the ability to support goal-directed reasoning by simulating, comparing, and selecting among alternative futures in both structured and open-ended environments. We build a task taxonomy and curate diverse datasets designed to probe these capabilities, moving beyond single-turn and perceptual evaluations. Through extensive experiments with state-of-the-art WMs, our results expose a substantial gap between current models and human-level hypothetical reasoning, and establish WR-Arena as both a diagnostic tool and a guideline for advancing next-generation world models capable of robust understanding, forecasting, and purposeful action. The code is available at https://github.com/MBZUAI-IFM/WR-Arena. |
|
| Do Language Models Follow Occam's Razor? An Evaluation of Parsimony in Inductive and Abductive Reasoning | 2026-03-26 | ShowNon-deductive reasoning, encompassing inductive and abductive reasoning, is essential in addressing complex real-world questions. One key feature of inductive and abductive reasoning is that there are many valid hypotheses; the simplest ones (those that adhere to Occam's Razor) are often most useful. However, this aspect is ignored in recent work that evaluates the non-deductive reasoning capabilities of large language models (LLMs). This work fills this gap, focusing on understanding whether the inductive and abductive reasoning capabilities of LLMs adhere to Occam's Razor, while also examining the correctness of their reasoning. To accomplish this goal, we introduce a framework to synthetically generate reasoning questions that (a) require inductive reasoning and abductive reasoning simultaneously; (b) is readily extended to produce any abductive/inductive reasoning question expressible in first-order logic. The task for the intelligent agent is to produce hypotheses to explain observations under a given world model. We also propose a new automated metric to assess whether hypotheses quantitatively adhere to Occam's Razor; those hypotheses that are correct and simplest are considered high-quality. Our findings on state-of-the-art LLMs suggest that LLMs can perform inductive and abductive reasoning in simple scenarios, but struggle with complex world models and with producing high-quality hypotheses, even with popular reasoning-enhancing techniques such as in-context learning and RLVR. |
|
| Persistent Robot World Models: Stabilizing Multi-Step Rollouts via Reinforcement Learning | 2026-03-26 | ShowAction-conditioned robot world models generate future video frames of the manipulated scene given a robot action sequence, offering a promising alternative for simulating tasks that are difficult to model with traditional physics engines. However, these models are optimized for short-term prediction and break down when deployed autoregressively: each predicted clip feeds back as context for the next, causing errors to compound and visual quality to rapidly degrade. We address this through the following contributions. First, we introduce a reinforcement learning (RL) post-training scheme that trains the world model on its own autoregressive rollouts rather than on ground-truth histories. We achieve this by adapting a recent contrastive RL objective for diffusion models to our setting and show that its convergence guarantees carry over exactly. Second, we design a training protocol that generates and compares multiple candidate variable-length futures from the same rollout state, reinforcing higher-fidelity predictions over lower-fidelity ones. Third, we develop efficient, multi-view visual fidelity rewards that combine complementary perceptual metrics across camera views and are aggregated at the clip level for dense, low-variance training signal. Fourth, we show that our approach establishes a new state-of-the-art for rollout fidelity on the DROID dataset, outperforming the strongest baseline on all metrics (e.g., LPIPS reduced by 14% on external cameras, SSIM improved by 9.1% on the wrist camera), winning 98% of paired comparisons, and achieving an 80% preference rate in a blind human study. |
34 pa...34 pages, 11 figures, 12 tables |
| Modernising Reinforcement Learning-Based Navigation for Embodied Semantic Scene Graph Generation | 2026-03-26 | ShowSemantic world models enable embodied agents to reason about objects, relations, and spatial context beyond purely geometric representations. In Organic Computing, such models are a key enabler for objective-driven self-adaptation under uncertainty and resource constraints. The core challenge is to acquire observations maximising model quality and downstream usefulness within a limited action budget. Semantic scene graphs (SSGs) provide a structured and compact representation for this purpose. However, constructing them within a finite action horizon requires exploration strategies that trade off information gain against navigation cost and decide when additional actions yield diminishing returns. This work presents a modular navigation component for Embodied Semantic Scene Graph Generation and modernises its decision-making by replacing the policy-optimisation method and revisiting the discrete action formulation. We study compact and finer-grained, larger discrete motion sets and compare a single-head policy over atomic actions with a factorised multi-head policy over action components. We evaluate curriculum learning and optional depth-based collision supervision, and assess SSG completeness, execution safety, and navigation behaviour. Results show that replacing the optimisation algorithm alone improves SSG completeness by 21% relative to the baseline under identical reward shaping. Depth mainly affects execution safety (collision-free motion), while completeness remains largely unchanged. Combining modern optimisation with a finer-grained, factorised action representation yields the strongest overall completeness--efficiency trade-off. |
|
| Lingshu-Cell: A generative cellular world model for transcriptome modeling toward virtual cells | 2026-03-26 | ShowModeling cellular states and predicting their responses to perturbations are central challenges in computational biology and the development of virtual cells. Existing foundation models for single-cell transcriptomics provide powerful static representations, but they do not explicitly model the distribution of cellular states for generative simulation. Here, we introduce Lingshu-Cell, a masked discrete diffusion model that learns transcriptomic state distributions and supports conditional simulation under perturbation. By operating directly in a discrete token space that is compatible with the sparse, non-sequential nature of single-cell transcriptomic data, Lingshu-Cell captures complex transcriptome-wide expression dependencies across approximately 18,000 genes without relying on prior gene selection, such as filtering by high variability or ranking by expression level. Across diverse tissues and species, Lingshu-Cell accurately reproduces transcriptomic distributions, marker-gene expression patterns and cell-subtype proportions, demonstrating its ability to capture complex cellular heterogeneity. Moreover, by jointly embedding cell type or donor identity with perturbation, Lingshu-Cell can predict whole-transcriptome expression changes for novel combinations of identity and perturbation. It achieves leading performance on the Virtual Cell Challenge H1 genetic perturbation benchmark and in predicting cytokine-induced responses in human PBMCs. Together, these results establish Lingshu-Cell as a flexible cellular world model for in silico simulation of cell states and perturbation responses, laying the foundation for a new paradigm in biological discovery and perturbation screening. |
|
| A Wireless World Model for AI-Native 6G Networks | 2026-03-26 | ShowIntegrating AI into the physical layer is a cornerstone of 6G networks. However, current data-driven approaches struggle to generalize across dynamic environments because they lack an intrinsic understanding of electromagnetic wave propagation. We introduce the Wireless World Model (WWM), a multi-modal foundation framework predicting the spatiotemporal evolution of wireless channels by internalizing the causal relationship between 3D geometry and signal dynamics. Pre-trained on a massive ray-traced multi-modal dataset, WWM overcomes the data authenticity gap, further validated under real-world measurement data. Using a joint-embedding predictive architecture with a multi-modal mixture-of-experts Transformer, WWM fuses channel state information, 3D point clouds, and user trajectories into a unified representation. Across the five key downstream tasks supported by WWM, it achieves remarkable performance in seen environments, unseen generalization scenarios, and real-world measurements, consistently outperforming SOTA uni-modal foundation models and task-specific models. This paves the way for physics-aware 6G intelligence that adapts to the physical world. |
|
| 3D Dynamics-Aware Manipulation: Endowing Manipulation Policies with 3D Foresight | 2026-03-26 | ShowThe incorporation of world modeling into manipulation policy learning has pushed the boundary of manipulation performance. However, existing efforts simply model the 2D visual dynamics, which is insufficient for robust manipulation when target tasks involve prominent depth-wise movement. To address this, we present a 3D dynamics-aware manipulation framework that seamlessly integrates 3D world modeling and policy learning. Three self-supervised learning tasks (current depth estimation, future RGB-D prediction, 3D flow prediction) are introduced within our framework, which complement each other and endow the policy model with 3D foresight. Extensive experiments on simulation and the real world show that 3D foresight can greatly boost the performance of manipulation policies without sacrificing inference speed. Code is available at https://github.com/Stardust-hyx/3D-Foresight. |
ICRA 2026 |
| NeoVerse: Enhancing 4D World Model with in-the-wild Monocular Videos | 2026-03-26 | ShowIn this paper, we propose NeoVerse, a versatile 4D world model that is capable of 4D reconstruction, novel-trajectory video generation, and rich downstream applications. We first identify a common limitation of scalability in current 4D world modeling methods, caused either by expensive and specialized multi-view 4D data or by cumbersome training pre-processing. In contrast, our NeoVerse is built upon a core philosophy that makes the full pipeline scalable to diverse in-the-wild monocular videos. Specifically, NeoVerse features pose-free feed-forward 4D reconstruction, online monocular degradation pattern simulation, and other well-aligned techniques. These designs empower NeoVerse with versatility and generalization to various domains. Meanwhile, NeoVerse achieves state-of-the-art performance in standard reconstruction and generation benchmarks. Our project page is available at https://neoverse-4d.github.io. |
CVPR ...CVPR 2026; Project Page: https://neoverse-4d.github.io |
| Unified Camera Positional Encoding for Controlled Video Generation | 2026-03-26 | ShowTransformers have emerged as a universal backbone across 3D perception, video generation, and world models for autonomous driving and embodied AI, where understanding camera geometry is essential for grounding visual observations in three-dimensional space. However, existing camera encoding methods often rely on simplified pinhole assumptions, restricting generalization across the diverse intrinsics and lens distortions in real-world cameras. We introduce Relative Ray Encoding, a geometry-consistent representation that unifies complete camera information, including 6-DoF poses, intrinsics, and lens distortions. To evaluate its capability under diverse controllability demands, we adopt camera-controlled text-to-video generation as a testbed task. Within this setting, we further identify pitch and roll as two components effective for Absolute Orientation Encoding, enabling full control over the initial camera orientation. Together, these designs form UCPE (Unified Camera Positional Encoding), which integrates into a pretrained video Diffusion Transformer through a lightweight spatial attention adapter, adding less than 1% trainable parameters while achieving state-of-the-art camera controllability and visual fidelity. To facilitate systematic training and evaluation, we construct a large video dataset covering a wide range of camera motions and lens types. Extensive experiments validate the effectiveness of UCPE in camera-controllable video generation and highlight its potential as a general camera representation for Transformers across future multi-view, video, and 3D tasks. Code will be available at https://github.com/chengzhag/UCPE. |
Camer...Camera Ready of CVPR2026. Project Page: https://chengzhag.github.io/publication/ucpe/ Code: https://github.com/chengzhag/UCPE |
| AI-Supervisor: Autonomous AI Research Supervision via a Persistent Research World Model | 2026-03-26 | ShowExisting automated research systems operate as stateless, linear pipelines -- generating outputs without maintaining any persistent understanding of the research landscape they navigate. They process papers sequentially, propose ideas without structured gap analysis, and lack mechanisms for agents to verify, challenge, or refine each other's findings. We present \textbf{AI-Supervisor}, a multi-agent orchestration framework where specialized agents provide end-to-end AI research supervision driven by human interests -- from literature review through gap discovery, method development, evaluation, and paper writing -- through autonomous exploration and self-correcting updates of research knowledge. Unlike sequential pipelines, AI-Supervisor maintains a continuously evolving \emph{Research World Model}, implemented as a Knowledge Graph, that captures methods, benchmarks, known limitations, and unexplored gaps, serving as shared memory across all agents and enabling agents to explore and build upon a structured understanding of the research landscape. The framework introduces three architectural contributions: (1) \emph{structured gap discovery} that decomposes methods into core modules, validates their performance across benchmarks, and maps the specific gaps each module creates; (2) \emph{self-correcting discovery loops} that probe why modules succeed on certain problems and fail on others, whether benchmarks carry hidden biases, and whether evaluation protocols remain adequate for emerging challenges; and (3) \emph{self-improving development loops} governed by cross-domain mechanism search that iteratively targets failing modules by finding solutions from other scientific fields. All agents operate under a \emph{consensus mechanism} where independent findings are corroborated before being committed to the Research World Model. |
|
| DCARL: A Divide-and-Conquer Framework for Autoregressive Long-Trajectory Video Generation | 2026-03-25 | ShowLong-trajectory video generation is a crucial yet challenging task for world modeling primarily due to the limited scalability of existing video diffusion models (VDMs). Autoregressive models, while offering infinite rollout, suffer from visual drift and poor controllability. To address these issues, we propose DCARL, a novel divide-and-conquer, autoregressive framework that effectively combines the structural stability of the divide-and-conquer scheme with the high-fidelity generation of VDMs. Our approach first employs a dedicated Keyframe Generator trained without temporal compression to establish long-range, globally consistent structural anchors. Subsequently, an Interpolation Generator synthesizes the dense frames in an autoregressive manner with overlapping segments, utilizing the keyframes for global context and a single clean preceding frame for local coherence. Trained on a large-scale internet long trajectory video dataset, our method achieves superior performance in both visual quality (lower FID and FVD) and camera adherence (lower ATE and ARE) compared to state-of-the-art autoregressive and divide-and-conquer baselines, demonstrating stable and high-fidelity generation for long trajectory videos up to 32 seconds in length. |
29 pa...29 pages, 11 figures. Project page: https://junyiouy.github.io/projects/dcarl |
| The Blueprints of Intelligence: A Functional-Topological Foundation for Perception and Representation | 2026-03-25 | ShowReal-world phenomena do not generate arbitrary variability: their signals concentrate on compact, low-variability subsets of functional space, enabling rapid generalization from few examples. A small child can recognize a dog after extremely limited exposure because the perceptual manifold of dog is compact, structured, and low-dimensional. We formalize this principle through a deterministic functional-topological framework in which the set of valid realizations produced by a physical process forms a compact subset of a Banach space, endowed with stable invariants, a finite Hausdorff radius, and an induced continuous perceptual functional. This geometry provides explicit limits on knowledge, conditions for identifiability, and guarantees for generalization from sparse evidence, properties fundamental to both natural and artificial intelligence. Across electromechanical, electrochemical, physiological, and multimodal cognitive domains (vision and speech), we show that real-world processes consistently generate compact perceptual manifolds with the same geometric characteristics. Their boundaries can be discovered in a fully self-supervised manner as the empirical radius saturates with increasing sampling, even when the governing equations are unknown. Crucially, we validate this framework empirically on the MNIST and Speech MNIST datasets: by treating classes as compact manifolds and using a purely topological proximity inference, the system achieves 90% accuracy with only 120 samples per class, directly mirroring the rapid geometric saturation predicted by the theory. These results demonstrate that deterministic functional topology offers a unified mathematical foundation for perception, representation, and world-model construction, establishing compact perceptual manifolds as a fundamental building block for future AI architectures. |
35 pa...35 pages, 6 figures. This preprint develops a deterministic functional-topological framework showing that physical systems generate compact perceptual manifolds with finite radius. We provide theory, Monte-Carlo estimators, and validation across PM, battery, and ECG domains, unifying biological perception and self-supervised AI |
| Latent-WAM: Latent World Action Modeling for End-to-End Autonomous Driving | 2026-03-25 | ShowWe introduce Latent-WAM, an efficient end-to-end autonomous driving framework that achieves strong trajectory planning through spatially-aware and dynamics-informed latent world representations. Existing world-model-based planners suffer from inadequately compressed representations, limited spatial understanding, and underutilized temporal dynamics, resulting in sub-optimal planning under constrained data and compute budgets. Latent-WAM addresses these limitations with two core modules: a Spatial-Aware Compressive World Encoder (SCWE) that distills geometric knowledge from a foundation model and compresses multi-view images into compact scene tokens via learnable queries, and a Dynamic Latent World Model (DLWM) that employs a causal Transformer to autoregressively predict future world status conditioned on historical visual and motion representations. Extensive experiments on NAVSIM v2 and HUGSIM demonstrate new state-of-the-art results: 89.3 EPDMS on NAVSIM v2 and 28.9 HD-Score on HUGSIM, surpassing the best prior perception-free method by 3.2 EPDMS with significantly less training data and a compact 104M-parameter model. |
|
| SEGAR: Selective Enhancement for Generative Augmented Reality | 2026-03-25 | ShowGenerative world models offer a compelling foundation for augmented-reality (AR) applications: by predicting future image sequences that incorporate deliberate visual edits, they enable temporally coherent, augmented future frames that can be computed ahead of time and cached, avoiding per-frame rendering from scratch in real time. In this work, we present SEGAR, a preliminary framework that combines a diffusion-based world model with a selective correction stage to support this vision. The world model generates augmented future frames with region-specific edits while preserving others, and the correction stage subsequently aligns safety-critical regions with real-world observations while preserving intended augmentations elsewhere. We demonstrate this pipeline in driving scenarios as a representative setting where semantic region structure is well defined and real-world feedback is readily available. We view this as an early step toward generative world models as practical AR infrastructure, where future frames can be generated, cached, and selectively corrected on demand. |
|
| FOCUS: Optimal Control for Multi-Entity World Modeling in Text-to-Image Generation | 2026-03-25 | ShowText-to-image (T2I) models excel on single-entity prompts but struggle with multi-entity scenes, often exhibiting attribute leakage, identity entanglement, and subject omissions. We present a principled theoretical framework that steers sampling toward multi-subject fidelity by casting flow matching (FM) as stochastic optimal control (SOC), yielding a single hyperparameter controlled trade-off between fidelity and object-centric state separation / binding consistency. Within this framework, we derive two architecture-agnostic algorithms: (i) a training-free test-time controller that perturbs the base velocity with a single-pass update, and (ii) Adjoint Matching, a lightweight fine-tuning rule that regresses a control network to a backward adjoint signal. The same formulation unifies prior attention heuristics, extends to diffusion models via a flow--diffusion correspondence, and provides the first fine-tuning route explicitly designed for multi-subject fidelity. In addition, we also introduce FOCUS (Flow Optimal Control for Unentangled Subjects), a probabilistic attention-binding objective compatible with both algorithms. Empirically, on Stable Diffusion 3.5 and FLUX.1, both algorithms consistently improve multi-subject alignment while maintaining base-model style; test-time control runs efficiently on commodity GPUs, and fine-tuned models generalize to unseen prompts. |
Proje...Project Page: https://ericbill21.github.io/FOCUS/ |
| CUA-Suite: Massive Human-annotated Video Demonstrations for Computer-Use Agents | 2026-03-25 | ShowComputer-use agents (CUAs) hold great promise for automating complex desktop workflows, yet progress toward general-purpose agents is bottlenecked by the scarcity of continuous, high-quality human demonstration videos. Recent work emphasizes that continuous video, not sparse screenshots, is the critical missing ingredient for scaling these agents. However, the largest existing open dataset, ScaleCUA, contains only 2 million screenshots, equating to less than 20 hours of video. To address this bottleneck, we introduce CUA-Suite, a large-scale ecosystem of expert video demonstrations and dense annotations for professional desktop computer-use agents. At its core is VideoCUA, which provides approximately 10,000 human-demonstrated tasks across 87 diverse applications with continuous 30 fps screen recordings, kinematic cursor traces, and multi-layerfed reasoning annotations, totaling approximately 55 hours and 6 million frames of expert video. Unlike sparse datasets that capture only final click coordinates, these continuous video streams preserve the full temporal dynamics of human interaction, forming a superset of information that can be losslessly transformed into the formats required by existing agent frameworks. CUA-Suite further provides two complementary resources: UI-Vision, a rigorous benchmark for evaluating grounding and planning capabilities in CUAs, and GroundCUA, a large-scale grounding dataset with 56K annotated screenshots and over 3.6 million UI element annotations. Preliminary evaluation reveals that current foundation action models struggle substantially with professional desktop applications (~60% task failure rate). Beyond evaluation, CUA-Suite's rich multimodal corpus supports emerging research directions including generalist screen parsing, continuous spatial control, video-based reward modeling, and visual world models. All data and models are publicly released. |
Proje...Project Page: https://cua-suite.github.io/ |
| GameplayQA: A Benchmarking Framework for Decision-Dense POV-Synced Multi-Video Understanding of 3D Virtual Agents | 2026-03-25 | ShowMultimodal LLMs are increasingly deployed as perceptual backbones for autonomous agents in 3D environments, from robotics to virtual worlds. These applications require agents to perceive rapid state changes, attribute actions to the correct entities, and reason about concurrent multi-agent behaviors from a first-person perspective, capabilities that existing benchmarks do not adequately evaluate. We introduce GameplayQA, a framework for evaluating agentic-centric perception and reasoning through video understanding. Specifically, we densely annotate multiplayer 3D gameplay videos at 1.22 labels/second, with time-synced, concurrent captions of states, actions, and events structured around a triadic system of Self, Other Agents, and the World, a natural decomposition for multi-agent environments. From these annotations, we refined 2.4K diagnostic QA pairs organized into three levels of cognitive complexity, accompanied by a structured distractor taxonomy that enables fine-grained analysis of where models hallucinate. Evaluation of frontier MLLMs reveals a substantial gap from human performance, with common failures in temporal and cross-video grounding, agent-role attribution, and handling the decision density of the game. We hope GameplayQA stimulates future research at the intersection of embodied AI, agentic perception, and world modeling. |
|
| LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels | 2026-03-24 | ShowJoint Embedding Predictive Architectures (JEPAs) offer a compelling framework for learning world models in compact latent spaces, yet existing methods remain fragile, relying on complex multi-term losses, exponential moving averages, pre-trained encoders, or auxiliary supervision to avoid representation collapse. In this work, we introduce LeWorldModel (LeWM), the first JEPA that trains stably end-to-end from raw pixels using only two loss terms: a next-embedding prediction loss and a regularizer enforcing Gaussian-distributed latent embeddings. This reduces tunable loss hyperparameters from six to one compared to the only existing end-to-end alternative. With ~15M parameters trainable on a single GPU in a few hours, LeWM plans up to 48x faster than foundation-model-based world models while remaining competitive across diverse 2D and 3D control tasks. Beyond control, we show that LeWM's latent space encodes meaningful physical structure through probing of physical quantities. Surprise evaluation confirms that the model reliably detects physically implausible events. |
|
| ManiDreams: An Open-Source Library for Robust Object Manipulation via Uncertainty-aware Task-specific Intuitive Physics | 2026-03-24 | ShowDynamics models, whether simulators or learned world models, have long been central to robotic manipulation, but most focus on minimizing prediction error rather than confronting a more fundamental challenge: real-world manipulation is inherently uncertain. We argue that robust manipulation under uncertainty is fundamentally an integration problem: uncertainties must be represented, propagated, and constrained within the planning loop, not merely suppressed during training. We present and open-source ManiDreams, a modular framework for uncertainty-aware manipulation planning over intuitive physics models. It realizes this integration through composable abstractions for distributional state representation, backend-agnostic dynamics prediction, and declarative constraint specification for action optimization. The framework explicitly addresses three sources of uncertainty: perceptual, parametric, and structural. It wraps any base policy with a sample-predict-constrain loop that evaluates candidate actions against distributional outcomes, adding robustness without retraining. Experiments on ManiSkill tasks show that ManiDreams maintains robust performance under various perturbations where the RL baseline degrades significantly. Runnable examples on pushing, picking, catching, and real-world deployment demonstrate flexibility across different policies, optimizers, physics backends, and executors. The framework is publicly available at https://github.com/Rice-RobotPI-Lab/ManiDreams |
9 pag...9 pages, 10 figures. Project page at https://rice-robotpi-lab.github.io/ManiDreams/ |
| WildWorld: A Large-Scale Dataset for Dynamic World Modeling with Actions and Explicit State toward Generative ARPG | 2026-03-24 | ShowDynamical systems theory and reinforcement learning view world evolution as latent-state dynamics driven by actions, with visual observations providing partial information about the state. Recent video world models attempt to learn this action-conditioned dynamics from data. However, existing datasets rarely match the requirement: they typically lack diverse and semantically meaningful action spaces, and actions are directly tied to visual observations rather than mediated by underlying states. As a result, actions are often entangled with pixel-level changes, making it difficult for models to learn structured world dynamics and maintain consistent evolution over long horizons. In this paper, we propose WildWorld, a large-scale action-conditioned world modeling dataset with explicit state annotations, automatically collected from a photorealistic AAA action role-playing game (Monster Hunter: Wilds). WildWorld contains over 108 million frames and features more than 450 actions, including movement, attacks, and skill casting, together with synchronized per-frame annotations of character skeletons, world states, camera poses, and depth maps. We further derive WildBench to evaluate models through Action Following and State Alignment. Extensive experiments reveal persistent challenges in modeling semantically rich actions and maintaining long-horizon state consistency, highlighting the need for state-aware video generation. The project page is https://shandaai.github.io/wildworld-project/. |
|
| VTAM: Video-Tactile-Action Models for Complex Physical Interaction Beyond VLAs | 2026-03-24 | ShowVideo-Action Models (VAMs) have emerged as a promising framework for embodied intelligence, learning implicit world dynamics from raw video streams to produce temporally consistent action predictions. Although such models demonstrate strong performance on long-horizon tasks through visual reasoning, they remain limited in contact-rich scenarios where critical interaction states are only partially observable from vision alone. In particular, fine-grained force modulation and contact transitions are not reliably encoded in visual tokens, leading to unstable or imprecise behaviors. To bridge this gap, we introduce the Video-Tactile Action Model (VTAM), a multimodal world modeling framework that incorporates tactile perception as a complementary grounding signal. VTAM augments a pretrained video transformer with tactile streams via a lightweight modality transfer finetuning, enabling efficient cross-modal representation learning without tactile-language paired data or independent tactile pretraining. To stabilize multimodal fusion, we introduce a tactile regularization loss that enforces balanced cross-modal attention, preventing visual latent dominance in the action model. VTAM demonstrates superior performance in contact-rich manipulation, maintaining a robust success rate of 90 percent on average. In challenging scenarios such as potato chip pick-and-place requiring high-fidelity force awareness, VTAM outperforms the pi 0.5 baseline by 80 percent. Our findings demonstrate that integrating tactile feedback is essential for correcting visual estimation errors in world action models, providing a scalable approach to physically grounded embodied foundation models. |
|
| EVA: Aligning Video World Models with Executable Robot Actions via Inverse Dynamics Rewards | 2026-03-24 | ShowVideo generative models are increasingly used as world models for robotics, where a model generates a future visual rollout conditioned on the current observation and task instruction, and an inverse dynamics model (IDM) converts the generated frames into executable robot actions. However, current video world models lack explicit executability constraints. As a result, visually coherent rollouts may still violate rigid-body and kinematic consistency, producing unstable or infeasible control commands when decoded by an IDM. We refer to this mismatch between visual generation and physically executable control as the executability gap. While this gap can be mitigated at inference time using techniques such as rejection sampling, such approaches are inefficient due to the high cost of video generation. In this paper, we leverage the executability gap as a training signal and introduce Executable Video Alignment (EVA), a reinforcement-learning post-training framework for aligning video world models. EVA trains an inverse dynamics model on real robot trajectories and repurposes it as a reward model that evaluates generated videos through the action sequences they induce, encouraging smooth motions measured by velocity, acceleration, and jerk while penalizing actions that violate embodiment constraints. Importantly, the reward remains informative even when generated videos contain severe visual artifacts, since such artifacts typically translate into unstable or out-of-bound actions. Experiments on the RoboTwin benchmark and a real bimanual robot show that EVA reduces embodiment-specific artifacts in generated rollouts and improves downstream task execution success. |
Proje...Project page: https://eva-project-page.github.io/ |
| Title | Date | Abstract | Comment |
|---|---|---|---|
| HyVGGT-VO: Tightly Coupled Hybrid Dense Visual Odometry with Feed-Forward Models | 2026-04-02 | ShowDense visual odometry (VO), which provides pose estimation and dense 3D reconstruction, serves as the cornerstone for applications ranging from robotics to augmented reality. Recently, feed-forward models have demonstrated remarkable capabilities in dense mapping. However, when these models are used in dense visual SLAM systems, their heavy computational burden restricts them to yielding sparse pose outputs at keyframes while still failing to achieve real-time pose estimation. In contrast, traditional sparse methods provide high computational efficiency and high-frequency pose outputs, but lack the capability for dense reconstruction. To address these limitations, we propose HyVGGT-VO, a novel framework that combines the computational efficiency of sparse VO with the dense reconstruction capabilities of feed-forward models. To the best of our knowledge, this is the first work to tightly couple a traditional VO framework with VGGT, a state-of-the-art feed-forward model. Specifically, we design an adaptive hybrid tracking frontend that dynamically switches between traditional optical flow and the VGGT tracking head to ensure robustness. Furthermore, we introduce a hierarchical optimization framework that jointly refines VO poses and the scale of VGGT predictions to ensure global scale consistency. Our approach achieves an approximately 5x processing speedup compared to existing VGGT-based methods, while reducing the average trajectory error by 85% on the indoor EuRoC dataset and 12% on the outdoor KITTI benchmark. Our code will be publicly available upon acceptance. Project page: https://geneta2580.github.io/HyVGGT-VO.io. |
|
| ToFormer: Towards Large-scale Scenario Depth Completion for Lightweight ToF Camera | 2026-03-21 | ShowTime-of-Flight (ToF) cameras possess compact design and high measurement precision to be applied to various robot tasks. However, their limited sensing range restricts deployment in large-scale scenarios. Depth completion has emerged as a potential solution to expand the sensing range of ToF cameras, but existing research lacks dedicated datasets and struggles to generalize to ToF measurements. In this paper, we propose a full-stack framework that enables depth completion in large-scale scenarios for short-range ToF cameras. First, we construct a multi-sensor platform with a reconstruction-based pipeline to collect real-world ToF samples with dense large-scale ground truth, yielding the first LArge-ScalE scenaRio ToF depth completion dataset (LASER-ToF). Second, we propose a sensor-aware depth completion network that incorporates a novel 3D branch with a 3D-2D Joint Propagation Pooling (JPP) module and Multimodal Cross-Covariance Attention (MXCA), enabling effective modeling of long-range relationships and efficient 3D-2D fusion under non-uniform ToF depth sparsity. Moreover, our network can utilize the sparse point cloud from visual SLAM as a supplement to ToF depth to further improve prediction accuracy. Experiments show that our method achieves an 8.6% lower mean absolute error than the second-best method, while maintaining lightweight design to support onboard deployment. Finally, to verify the system's applicability on real robots, we deploy proposed method on a quadrotor at a 10Hz runtime, enabling reliable large-scale mapping and long-range planning in challenging environments for short-range ToF cameras. |
17 pages, 15 figures |
| PathSpace: Rapid continuous map approximation for efficient SLAM using B-Splines in constrained environments | 2026-03-19 | ShowSimultaneous Localization and Mapping (SLAM) plays a crucial role in enabling autonomous vehicles to navigate previously unknown environments. Semantic SLAM mostly extends visual SLAM, leveraging the higher density information available to reason about the environment in a more human-like manner. This allows for better decision making by exploiting prior structural knowledge of the environment, usually in the form of labels. Current semantic SLAM techniques still mostly rely on a dense geometric representation of the environment, limiting their ability to apply constraints based on context. We propose PathSpace, a novel semantic SLAM framework that uses continuous B-splines to represent the environment in a compact manner, while also maintaining and reasoning through the continuous probability density functions required for probabilistic reasoning. This system applies the multiple strengths of B-splines in the context of SLAM to interpolate and fit otherwise discrete sparse environments. We test this framework in the context of autonomous racing, where we exploit pre-specified track characteristics to produce significantly reduced representations at comparable levels of accuracy to traditional landmark based methods and demonstrate its potential in limiting the resources used by a system with minimal accuracy loss. |
|
| Visual SLAM with DEM Anchoring for Lunar Surface Navigation | 2026-03-18 | ShowFuture lunar missions will require autonomous rovers capable of traversing tens of kilometers across challenging terrain while maintaining accurate localization and producing globally consistent maps. However, the absence of global positioning systems, extreme illumination, and low-texture regolith make long-range navigation on the Moon particularly difficult, as visual-inertial odometry pipelines accumulate drift over extended traverses. To address this challenge, we present a stereo visual simultaneous localization and mapping (SLAM) system that integrates learned feature detection and matching with global constraints from digital elevation models (DEMs). Our front-end employs learning-based feature extraction and matching to achieve robustness to illumination extremes and repetitive terrain, while the back-end incorporates DEM-derived height and surface-normal factors into a pose graph, providing absolute surface constraints that mitigate long-term drift. We validate our approach using both simulated lunar traverse data generated in Unreal Engine and real Moon/Mars analog data collected from Mt. Etna. Results demonstrate that DEM anchoring consistently reduces absolute trajectory error compared to baseline SLAM methods, lowering drift in long-range navigation even in repetitive or visually aliased terrain. |
Accep...Accepted to IEEE Aerospace Conference 2026 |
| FastLoop: Parallel Loop Closing with GPU-Acceleration in Visual SLAM | 2026-03-17 | ShowVisual SLAM systems combine visual tracking with global loop closure to maintain a consistent map and accurate localization. Loop closure is a computationally expensive process as we need to search across the whole map for matches. This paper presents FastLoop, a GPU-accelerated loop closing module to alleviate this computational complexity. We identify key performance bottlenecks in the loop closing pipeline of visual SLAM and address them through parallel optimizations on the GPU. Specifically, we use task-level and data-level parallelism and integrate a GPU-accelerated pose graph optimization. Our implementation is built on top of ORB-SLAM3 and leverages CUDA for GPU programming. Experimental results show that FastLoop achieves an average speedup of 1.4x and 1.3x on the EuRoC dataset and 3.0x and 2.4x on the TUM-VI dataset for the loop closing module on desktop and embedded platforms, respectively, while maintaining the accuracy of the original system. |
|
| SLAM Adversarial Lab: An Extensible Framework for Visual SLAM Robustness Evaluation under Adverse Conditions | 2026-03-17 | ShowWe present SAL (SLAM Adversarial Lab), a modular framework for evaluating visual SLAM systems under adversarial conditions such as fog and rain. SAL represents each adversarial condition as a perturbation that transforms an existing dataset into an adversarial dataset. When transforming a dataset, SAL supports severity levels using easily-interpretable real-world units such as meters for fog visibility. SAL's extensible architecture decouples datasets, perturbations, and SLAM algorithms through common interfaces, so users can add new components without rewriting integration code. Moreover, SAL includes a search procedure that finds the severity level of a perturbation at which a SLAM system fails. To showcase the capabilities of SAL, our evaluation integrates seven SLAM algorithms and evaluates them across three datasets under weather, camera, and video transport perturbations. |
8 pages, 4 figures |
| TurboMap: GPU-Accelerated Local Mapping for Visual SLAM | 2026-03-17 | ShowIn real-time Visual SLAM systems, local mapping must operate under strict latency constraints, as delays degrade map quality and increase the risk of tracking failure. GPU parallelization offers a promising way to reduce latency. However, parallelizing local mapping is challenging due to synchronized shared-state updates and the overhead of transferring large map data structures to the GPU. This paper presents TurboMap, a GPU-parallelized and CPU-optimized local mapping backend that holistically addresses these challenges. We restructure Map Point Creation to enable parallel Keypoint Correspondence Search on the GPU, redesign and parallelize Map Point Fusion, optimize Redundant Keyframe Culling on the CPU, and integrate a fast GPU-based Local Bundle Adjustment solver. To minimize data transfer and synchronization costs, we introduce persistent GPU-resident keyframe storage. Experiments on the EuRoC and TUM-VI datasets show average local mapping speedups of 1.3x and 1.6x, respectively, while preserving accuracy. |
|
| Industrial cuVSLAM Benchmark & Integration | 2026-03-17 | ShowThis work presents a comprehensive benchmark evaluation of visual odometry (VO) and visual SLAM (VSLAM) systems for mobile robot navigation in real-world logistical environments. We compare multiple visual odometry approaches across controlled trajectories covering translational, rotational, and mixed motion patterns, as well as a large-scale production facility dataset spanning approximately 1.7 km. Performance is evaluated using Absolute Pose Error (APE) against ground truth from a Vicon motion capture system and a LiDAR-based SLAM reference. Our results show that a hybrid stack combining the cuVSLAM front-end with a custom SLAM back-end achieves the strongest mapping accuracy, motivating a deeper integration of cuVSLAM as the core VO component in our robotics stack. We further validate this integration by deploying and testing the cuVSLAM-based VO stack on an NVIDIA Jetson platform. |
|
| Dense Dynamic Scene Reconstruction and Camera Pose Estimation from Multi-View Videos | 2026-03-14 | ShowWe address the challenging problem of dense dynamic scene reconstruction and camera pose estimation from multiple freely moving cameras -- a setting that arises naturally when multiple observers capture a shared event. Prior approaches either handle only single-camera input or require rigidly mounted, pre-calibrated camera rigs, limiting their practical applicability. We propose a two-stage optimization framework that decouples the task into robust camera tracking and dense depth refinement. In the first stage, we extend single-camera visual SLAM to the multi-camera setting by constructing a spatiotemporal connection graph that exploits both intra-camera temporal continuity and inter-camera spatial overlap, enabling consistent scale and robust tracking. To ensure robustness under limited overlap, we introduce a wide-baseline initialization strategy using feed-forward reconstruction models. In the second stage, we refine depth and camera poses by optimizing dense inter- and intra-camera consistency using wide-baseline optical flow. Additionally, we introduce MultiCamRobolab, a new real-world dataset with ground-truth poses from a motion capture system. Finally, we demonstrate that our method significantly outperforms state-of-the-art feed-forward models on both synthetic and real-world benchmarks, while requiring less memory. |
fix typos |
| vS-Graphs: Tightly Coupling Visual SLAM and 3D Scene Graphs Exploiting Hierarchical Scene Understanding | 2026-03-11 | ShowCurrent Visual Simultaneous Localization and Mapping (VSLAM) systems often struggle to create maps that are both semantically rich and easily interpretable. While incorporating semantic scene knowledge aids in building richer maps with contextual associations among mapped objects, representing them in structured formats, such as scene graphs, has not been widely addressed, resulting in complex map comprehension and limited scalability. This paper introduces vS-Graphs, a novel real-time VSLAM framework that integrates vision-based scene understanding with map reconstruction and comprehensible graph-based representation. The framework infers structural elements (i.e., rooms and floors) from detected building components (i.e., walls and ground surfaces) and incorporates them into optimizable 3D scene graphs. This solution enhances the reconstructed map's semantic richness, comprehensibility, and localization accuracy. Extensive experiments on standard benchmarks and real-world datasets demonstrate that vS-Graphs achieves an average of 15.22% accuracy gain across all tested datasets compared to state-of-the-art VSLAM methods. Furthermore, the proposed framework achieves environment-driven semantic entity detection accuracy comparable to that of precise LiDAR-based frameworks, using only visual features. The code is publicly available at https://github.com/snt-arg/visual_sgraphs and is actively being improved. Moreover, a web page containing more media and evaluation outcomes is available on https://snt-arg.github.io/vsgraphs-results/. |
20 pa...20 pages, 10 figures, 5 tables |
| LIVE-GS: Online LiDAR-Inertial-Visual State Estimation and Globally Consistent Mapping with 3D Gaussian Splatting | 2026-03-09 | ShowWhile 3D Gaussian Splatting (3DGS) enabled photorealistic mapping, its integration into SLAM has largely followed traditional camera-centric pipelines. As a result, they inherit well-known weaknesses such as high computational load, failure in texture-poor or illumination-varying environments, and limited operational range, particularly for RGB-D setups. On the other hand, LiDAR emerges as a robust alternative, but its integration with 3DGS introduces new challenges, such as the need for tighter global alignment for photorealistic quality and prolonged optimization times caused by sparse data. To address these challenges, we propose LIVE-GS, an online LiDAR-Inertial Visual SLAM framework that tightly couples 3D Gaussian Splatting with LiDAR-based surfels to ensure high-precision map consistency through global geometric optimization. Particularly, to handle sparse data, our system employs a depth-invariant Gaussian initialization strategy for efficient representation and a bounded sigmoid constraint to prevent uncontrolled Gaussian growth. Experiments on public and our datasets demonstrate competitive performance in rendering quality and map-building efficiency compared with representative 3DGS SLAM baselines. |
|
| Have We Mastered Scale in Deep Monocular Visual SLAM? The ScaleMaster Dataset and Benchmark | 2026-02-20 | ShowRecent advances in deep monocular visual Simultaneous Localization and Mapping (SLAM) have achieved impressive accuracy and dense reconstruction capabilities, yet their robustness to scale inconsistency in large-scale indoor environments remains largely unexplored. Existing benchmarks are limited to room-scale or structurally simple settings, leaving critical issues of intra-session scale drift and inter-session scale ambiguity insufficiently addressed. To fill this gap, we introduce the ScaleMaster Dataset, the first benchmark explicitly designed to evaluate scale consistency under challenging scenarios such as multi-floor structures, long trajectories, repetitive views, and low-texture regions. We systematically analyze the vulnerability of state-of-the-art deep monocular visual SLAM systems to scale inconsistency, providing both quantitative and qualitative evaluations. Crucially, our analysis extends beyond traditional trajectory metrics to include a direct map-to-map quality assessment using metrics like Chamfer distance against high-fidelity 3D ground truth. Our results reveal that while recent deep monocular visual SLAM systems demonstrate strong performance on existing benchmarks, they suffer from severe scale-related failures in realistic, large-scale indoor environments. By releasing the ScaleMaster dataset and baseline results, we aim to establish a foundation for future research toward developing scale-consistent and reliable visual SLAM systems. |
8 pag...8 pages, 9 figures, accepted to ICRA 2026 |
| Cholec80-port: A Geometrically Consistent Trocar Port Segmentation Dataset for Robust Surgical Scene Understanding | 2026-02-19 | ShowTrocar ports are camera-fixed, pseudo-static structures that can persistently occlude laparoscopic views and attract disproportionate feature points due to specular, textured surfaces. This makes ports particularly detrimental to geometry-based downstream pipelines such as image stitching, 3D reconstruction, and visual SLAM, where dynamic or non-anatomical outliers degrade alignment and tracking stability. Despite this practical importance, explicit port labels are rare in public surgical datasets, and existing annotations often violate geometric consistency by masking the central lumen (opening), even when anatomical regions are visible through it. We present Cholec80-port, a high-fidelity trocar port segmentation dataset derived from Cholec80, together with a rigorous standard operating procedure (SOP) that defines a port-sleeve mask excluding the central opening. We additionally cleanse and unify existing public datasets under the same SOP. Experiments demonstrate that geometrically consistent annotations substantially improve cross-dataset robustness beyond what dataset size alone provides. |
|
| Adaptive Illumination Control for Robot Perception | 2026-02-13 | ShowRobot perception under low light or high dynamic range is usually improved downstream - via more robust feature extraction, image enhancement, or closed-loop exposure control. However, all of these approaches are limited by the image captured these conditions. An alternate approach is to utilize a programmable onboard light that adds to ambient illumination and improves captured images. However, it is not straightforward to predict its impact on image formation. Illumination interacts nonlinearly with depth, surface reflectance, and scene geometry. It can both reveal structure and induce failure modes such as specular highlights and saturation. We introduce Lightning, a closed-loop illumination-control framework for visual SLAM that combines relighting, offline optimization, and imitation learning. This is performed in three stages. First, we train a Co-Located Illumination Decomposition (CLID) relighting model that decomposes a robot observation into an ambient component and a light-contribution field. CLID enables physically consistent synthesis of the same scene under alternative light intensities and thereby creates dense multi-intensity training data without requiring us to repeatedly re-run trajectories. Second, using these synthesized candidates, we formulate an offline Optimal Intensity Schedule (OIS) problem that selects illumination levels over a sequence trading off SLAM-relevant image utility against power consumption and temporal smoothness. Third, we distill this ideal solution into a real-time controller through behavior cloning, producing an Illumination Control Policy (ILC) that generalizes beyond the initial training distribution and runs online on a mobile robot to command discrete light-intensity levels. Across our evaluation, Lightning substantially improves SLAM trajectory robustness while reducing unnecessary illumination power. |
|
| Thegra: Graph-based SLAM for Thermal Imagery | 2026-02-09 | ShowThermal imaging provides a practical sensing modality for visual SLAM in visually degraded environments such as low illumination, smoke, or adverse weather. However, thermal imagery often exhibits low texture, low contrast, and high noise, complicating feature-based SLAM. In this work, we propose a sparse monocular graph-based SLAM system for thermal imagery that leverages general-purpose learned features -- the SuperPoint detector and LightGlue matcher, trained on large-scale visible-spectrum data to improve cross-domain generalization. To adapt these components to thermal data, we introduce a preprocessing pipeline to enhance input suitability and modify core SLAM modules to handle sparse and outlier-prone feature matches. We further incorporate keypoint confidence scores from SuperPoint into a confidence-weighted factor graph to improve estimation robustness. Evaluations on public thermal datasets demonstrate that the proposed system achieves reliable performance without requiring dataset-specific training or fine-tuning a desired feature detector, given the scarcity of quality thermal data. Code will be made available upon publication. |
|
| Real-Time Loop Closure Detection in Visual SLAM via NetVLAD and Faiss | 2026-02-02 | ShowLoop closure detection (LCD) is a core component of simultaneous localization and mapping (SLAM): it identifies revisited places and enables pose-graph constraints that correct accumulated drift. Classic bag-of-words approaches such as DBoW are efficient but often degrade under appearance change and perceptual aliasing. In parallel, deep learning-based visual place recognition (VPR) descriptors (e.g., NetVLAD and Transformer-based models) offer stronger robustness, but their computational cost is often viewed as a barrier to real-time SLAM. In this paper, we empirically evaluate NetVLAD as an LCD module and compare it against DBoW on the KITTI dataset. We introduce a Fine-Grained Top-K precision-recall curve that better reflects LCD settings where a query may have zero or multiple valid matches. With Faiss-accelerated nearestneighbor search, NetVLAD achieves real-time query speed while improving accuracy and robustness over DBoW, making it a practical drop-in alternative for LCD in SLAM. |
|
| When Simultaneous Localization and Mapping Meets Wireless Communications: A Survey | 2026-01-28 | ShowThe availability of commercial wireless communication and sensing equipment combined with the advancements in intelligent autonomous systems paves the way towards robust joint communications and simultaneous localization and mapping (SLAM). This paper surveys the state-of-the-art in the nexus of SLAM and Wireless Communications, attributing the bidirectional impact of each with a focus on visual SLAM (V-SLAM) integration. We provide an overview of key concepts related to wireless signal propagation, geometric channel modeling, and radio frequency (RF)-based localization and sensing. In addition to this, we show image processing techniques that can detect landmarks, proactively predicting optimal paths for wireless channels. Several dimensions are considered, including the prerequisites, techniques, background, and future directions and challenges of the intersection between SLAM and wireless communications. We analyze mathematical approaches such as probabilistic models, and spatial methods for signal processing, as well as key technological aspects. We expose techniques and items towards enabling a highly effective retrieval of the autonomous robot state. Among other interesting findings, we observe that monocular V-SLAM would benefit from RF relevant information, as the latter can serve as a proxy for the scale ambiguity resolution. Conversely, we find that wireless communications in the context of 5G and beyond can potentially benefit from visual odometry that is central in SLAM. Moreover, we examine other sources besides the camera for SLAM and describe the twofold relation with wireless communications. Finally, integrated solutions performing joint communications and SLAM are still in their infancy: theoretical and practical advancements are required to add higher-level localization and semantic perception capabilities to RF and multi-antenna technologies. |
|
| SCE-SLAM: Scale-Consistent Monocular SLAM via Scene Coordinate Embeddings | 2026-01-14 | ShowMonocular visual SLAM enables 3D reconstruction from internet video and autonomous navigation on resource-constrained platforms, yet suffers from scale drift, i.e., the gradual divergence of estimated scale over long sequences. Existing frame-to-frame methods achieve real-time performance through local optimization but accumulate scale drift due to the lack of global constraints among independent windows. To address this, we propose SCE-SLAM, an end-to-end SLAM system that maintains scale consistency through scene coordinate embeddings, which are learned patch-level representations encoding 3D geometric relationships under a canonical scale reference. The framework consists of two key modules: geometry-guided aggregation that leverages 3D spatial proximity to propagate scale information from historical observations through geometry-modulated attention, and scene coordinate bundle adjustment that anchors current estimates to the reference scale through explicit 3D coordinate constraints decoded from the scene coordinate embeddings. Experiments on KITTI, Waymo, and vKITTI demonstrate substantial improvements: our method reduces absolute trajectory error by 8.36m on KITTI compared to the best prior approach, while maintaining 36 FPS and achieving scale consistency across large-scale scenes. |
|
| VPGS-SLAM: Voxel-based Progressive 3D Gaussian SLAM in Large-Scale Scenes | 2026-01-10 | Show3D Gaussian Splatting has recently shown promising results in dense visual SLAM. However, existing 3DGS-based SLAM methods are all constrained to small-room scenarios and struggle with memory explosion in large-scale scenes and long sequences. To this end, we propose VPGS-SLAM, the first 3DGS-based large-scale RGBD SLAM framework for both indoor and outdoor scenarios. We design a novel voxel-based progressive 3D Gaussian mapping method with multiple submaps for compact and accurate scene representation in large-scale and long-sequence scenes. This allows us to scale up to arbitrary scenes and improves robustness (even under pose drifts). In addition, we propose a 2D-3D fusion camera tracking method to achieve robust and accurate camera tracking in both indoor and outdoor large-scale scenes. Furthermore, we design a 2D-3D Gaussian loop closure method to eliminate pose drift. We further propose a submap fusion method with online distillation to achieve global consistency in large-scale scenes when detecting a loop. Experiments on various indoor and outdoor datasets demonstrate the superiority and generalizability of the proposed framework. The code will be open source on https://github.com/dtc111111/vpgs-slam. |
|
| ViSTA-SLAM: Visual SLAM with Symmetric Two-view Association | 2026-01-06 | ShowWe present ViSTA-SLAM as a real-time monocular visual SLAM system that operates without requiring camera intrinsics, making it broadly applicable across diverse camera setups. At its core, the system employs a lightweight symmetric two-view association (STA) model as the frontend, which simultaneously estimates relative camera poses and regresses local pointmaps from only two RGB images. This design reduces model complexity significantly, the size of our frontend is only 35% that of comparable state-of-the-art methods, while enhancing the quality of two-view constraints used in the pipeline. In the backend, we construct a specially designed Sim(3) pose graph that incorporates loop closures to address accumulated drift. Extensive experiments demonstrate that our approach achieves superior performance in both camera tracking and dense 3D reconstruction quality compared to current methods. Github repository: https://github.com/zhangganlin/vista-slam |
Accep...Accepted by 3DV 2026, project page: https://ganlinzhang.xyz/vista-slam/ |
| Loop Closure using AnyLoc Visual Place Recognition in DPV-SLAM | 2026-01-06 | ShowLoop closure is crucial for maintaining the accuracy and consistency of visual SLAM. We propose a method to improve loop closure performance in DPV-SLAM. Our approach integrates AnyLoc, a learning-based visual place recognition technique, as a replacement for the classical Bag of Visual Words (BoVW) loop detection method. In contrast to BoVW, which relies on handcrafted features, AnyLoc utilizes deep feature representations, enabling more robust image retrieval across diverse viewpoints and lighting conditions. Furthermore, we propose an adaptive mechanism that dynamically adjusts similarity threshold based on environmental conditions, removing the need for manual tuning. Experiments on both indoor and outdoor datasets demonstrate that our method significantly outperforms the original DPV-SLAM in terms of loop closure accuracy and robustness. The proposed method offers a practical and scalable solution for enhancing loop closure performance in modern SLAM systems. |
Accep...Accepted at IEEE/SICE International Symposium on System Integration(SII) 2026. 6 pages, 14 figures |
| FoundationSLAM: Unleashing the Power of Depth Foundation Models for End-to-End Dense Visual SLAM | 2026-01-01 | ShowWe present FoundationSLAM, a learning-based monocular dense SLAM system that addresses the absence of geometric consistency in previous flow-based approaches for accurate and robust tracking and mapping. Our core idea is to bridge flow estimation with geometric reasoning by leveraging the guidance from foundation depth models. To this end, we first develop a Hybrid Flow Network that produces geometry-aware correspondences, enabling consistent depth and pose inference across diverse keyframes. To enforce global consistency, we propose a Bi-Consistent Bundle Adjustment Layer that jointly optimizes keyframe pose and depth under multi-view constraints. Furthermore, we introduce a Reliability-Aware Refinement mechanism that dynamically adapts the flow update process by distinguishing between reliable and uncertain regions, forming a closed feedback loop between matching and optimization. Extensive experiments demonstrate that FoundationSLAM achieves superior trajectory accuracy and dense reconstruction quality across multiple challenging datasets, while running in real-time at 18 FPS, demonstrating strong generalization to various scenarios and practical applicability of our method. |
Accep...Accept at AAAI 2026 (Oral) |
| Spatia: Video Generation with Updatable Spatial Memory | 2025-12-17 | ShowExisting video generation models struggle to maintain long-term spatial and temporal consistency due to the dense, high-dimensional nature of video signals. To overcome this limitation, we propose Spatia, a spatial memory-aware video generation framework that explicitly preserves a 3D scene point cloud as persistent spatial memory. Spatia iteratively generates video clips conditioned on this spatial memory and continuously updates it through visual SLAM. This dynamic-static disentanglement design enhances spatial consistency throughout the generation process while preserving the model's ability to produce realistic dynamic entities. Furthermore, Spatia enables applications such as explicit camera control and 3D-aware interactive editing, providing a geometrically grounded framework for scalable, memory-driven video generation. |
Proje...Project page: https://zhaojingjing713.github.io/Spatia/ |
| Deep Learning Perspective of Scene Understanding in Autonomous Robots | 2025-12-16 | ShowThis paper provides a review of deep learning applications in scene understanding in autonomous robots, including innovations in object detection, semantic and instance segmentation, depth estimation, 3D reconstruction, and visual SLAM. It emphasizes how these techniques address limitations of traditional geometric models, improve depth perception in real time despite occlusions and textureless surfaces, and enhance semantic reasoning to understand the environment better. When these perception modules are integrated into dynamic and unstructured environments, they become more effective in decisionmaking, navigation and interaction. Lastly, the review outlines the existing problems and research directions to advance learning-based scene understanding of autonomous robots. |
11 pa...11 pages. Review Paper on Deep Learning Perspective of Scene Understanding in Autonomous Robots |
| Dynamic Visual SLAM using a General 3D Prior | 2025-12-07 | ShowReliable incremental estimation of camera poses and 3D reconstruction is key to enable various applications including robotics, interactive visualization, and augmented reality. However, this task is particularly challenging in dynamic natural environments, where scene dynamics can severely deteriorate camera pose estimation accuracy. In this work, we propose a novel monocular visual SLAM system that can robustly estimate camera poses in dynamic scenes. To this end, we leverage the complementary strengths of geometric patch-based online bundle adjustment and recent feed-forward reconstruction models. Specifically, we propose a feed-forward reconstruction model to precisely filter out dynamic regions, while also utilizing its depth prediction to enhance the robustness of the patch-based visual SLAM. By aligning depth prediction with estimated patches from bundle adjustment, we robustly handle the inherent scale ambiguities of the batch-wise application of the feed-forward reconstruction model. |
8 pages |
| DPVO-QAT++: Heterogeneous QAT and CUDA Kernel Fusion for High-Performance Deep Patch Visual Odometry | 2025-11-16 | ShowDeep learning-based Visual SLAM (vSLAM) systems exhibit exceptional geometric reasoning capabilities, yet their prohibitive computational overhead severely restricts deployment on resource-constrained autonomous platforms. This paper presents a hierarchical quantization optimization framework, DPVO-QAT++ (DPVO-QAT++: Heterogeneous QAT and CUDA Kernel Fusion for High-Performance Deep Patch Visual Odometry). Through the synergistic integration of learnable scale parameterization, a heterogeneous precision design for the Visual Odometry (VO) front-end and back-end (front-end floating-point fake quantization with FP16/FP32; back-end full precision), and GPU-native kernel fusion for fake quantization (custom CUDA kernels), our framework significantly reduces memory footprint and increases processing speed while preserving the trajectory accuracy of the original model. On the TartanAir dataset, our framework achieves an average FPS increase of 52.1%, a 29.1% reduction in median latency, and a 64.9% reduction in peak GPU memory reservation, while maintaining trajectory accuracy (ATE) comparable to the original DPVO model across 32 validation sequences. On the EuRoC dataset, it realizes an average FPS increase of 30.1%, a 23.1% reduction in median latency, and a 37.7% reduction in peak GPU memory reservation, maintaining comparable trajectory accuracy (ATE) across 11 validation sequences. Experimental results demonstrate that DPVO-QAT++ effectively bridges the gap between high-precision deep VO and the efficiency requirements for practical deployment, offering a viable engineering paradigm for the application of this technology on real-world embedded platforms. Keywords: Visual Odometry, Heterogeneous Precision Architecture, Quantization-Aware Training, CUDA Kernel Fusion, Scale-Only Training, Deep Patch Visual Odometry, GPU-Native Kernel Fusion. |
|
| MASt3R-Fusion: Integrating Feed-Forward Visual Model with IMU, GNSS for High-Functionality SLAM | 2025-11-16 | ShowVisual SLAM is a cornerstone technique in robotics, autonomous driving and extended reality (XR), yet classical systems often struggle with low-texture environments, scale ambiguity, and degraded performance under challenging visual conditions. Recent advancements in feed-forward neural network-based pointmap regression have demonstrated the potential to recover high-fidelity 3D scene geometry directly from images, leveraging learned spatial priors to overcome limitations of traditional multi-view geometry methods. However, the widely validated advantages of probabilistic multi-sensor information fusion are often discarded in these pipelines. In this work, we propose MASt3R-Fusion,a multi-sensor-assisted visual SLAM framework that tightly integrates feed-forward pointmap regression with complementary sensor information, including inertial measurements and GNSS data. The system introduces Sim(3)-based visualalignment constraints (in the Hessian form) into a universal metric-scale SE(3) factor graph for effective information fusion. A hierarchical factor graph design is developed, which allows both real-time sliding-window optimization and global optimization with aggressive loop closures, enabling real-time pose tracking, metric-scale structure perception and globally consistent mapping. We evaluate our approach on both public benchmarks and self-collected datasets, demonstrating substantial improvements in accuracy and robustness over existing visual-centered multi-sensor SLAM systems. The code will be released open-source to support reproducibility and further research (https://github.com/GREAT-WHU/MASt3R-Fusion). |
|
| UMIGen: A Unified Framework for Egocentric Point Cloud Generation and Cross-Embodiment Robotic Imitation Learning | 2025-11-12 | ShowData-driven robotic learning faces an obvious dilemma: robust policies demand large-scale, high-quality demonstration data, yet collecting such data remains a major challenge owing to high operational costs, dependence on specialized hardware, and the limited spatial generalization capability of current methods. The Universal Manipulation Interface (UMI) relaxes the strict hardware requirements for data collection, but it is restricted to capturing only RGB images of a scene and omits the 3D geometric information on which many tasks rely. Inspired by DemoGen, we propose UMIGen, a unified framework that consists of two key components: (1) Cloud-UMI, a handheld data collection device that requires no visual SLAM and simultaneously records point cloud observation-action pairs; and (2) a visibility-aware optimization mechanism that extends the DemoGen pipeline to egocentric 3D observations by generating only points within the camera's field of view. These two components enable efficient data generation that aligns with real egocentric observations and can be directly transferred across different robot embodiments without any post-processing. Experiments in both simulated and real-world settings demonstrate that UMIGen supports strong cross-embodiment generalization and accelerates data collection in diverse manipulation tasks. |
|
| Integration of Visual SLAM into Consumer-Grade Automotive Localization | 2025-11-10 | ShowAccurate ego-motion estimation in consumer-grade vehicles currently relies on proprioceptive sensors, i.e. wheel odometry and IMUs, whose performance is limited by systematic errors and calibration. While visual-inertial SLAM has become a standard in robotics, its integration into automotive ego-motion estimation remains largely unexplored. This paper investigates how visual SLAM can be integrated into consumer-grade vehicle localization systems to improve performance. We propose a framework that fuses visual SLAM with a lateral vehicle dynamics model to achieve online gyroscope calibration under realistic driving conditions. Experimental results demonstrate that vision-based integration significantly improves gyroscope calibration accuracy and thus enhances overall localization performance, highlighting a promising path toward higher automotive localization accuracy. We provide results on both proprietary and public datasets, showing improved performance and superior localization accuracy on a public benchmark compared to state-of-the-art methods. |
This ...This manuscript has been submitted to the IEEE for possible publication |
| Multi-cam Multi-map Visual Inertial Localization: System, Validation and Dataset | 2025-11-08 | ShowRobot control loops require causal pose estimates that depend only on past and present measurements. At each timestep, controllers compute commands using the current pose without waiting for future refinements. While traditional visual SLAM systems achieve high accuracy through retrospective loop closures, these corrections arrive after control decisions were already executed, violating causality. Visual-inertial odometry maintains causality but accumulates unbounded drift over time. To address the distinct requirements of robot control, we propose a multi-camera multi-map visual-inertial localization system providing real-time, causal pose estimation with bounded localization error through continuous map constraints. Since standard trajectory metrics evaluate post-processed trajectories, we analyze the error composition of map-based localization systems and propose a set of evaluation metrics suitable for measuring causal localization performance. To validate our system, we design a multi-camera IMU hardware setup and collect a challenging long-term campus dataset featuring diverse illumination and seasonal conditions. Experimental results on public benchmarks and on our own collected dataset demonstrate that our system provides significantly higher real-time localization accuracy compared to other methods. To benefit the community, we have made both the system and the dataset open source at https://anonymous.4open.science/r/Multi-cam-Multi-map-VILO-7993. |
|
| Loop Closure from Two Views: Revisiting PGO for Scalable Trajectory Estimation through Monocular Priors | 2025-10-30 | Show(Visual) Simultaneous Localization and Mapping (SLAM) remains a fundamental challenge in enabling autonomous systems to navigate and understand large-scale environments. Traditional SLAM approaches struggle to balance efficiency and accuracy, particularly in large-scale settings where extensive computational resources are required for scene reconstruction and Bundle Adjustment (BA). However, this scene reconstruction, in the form of sparse pointclouds of visual landmarks, is often only used within the SLAM system because navigation and planning methods require different map representations. In this work, we therefore investigate a more scalable Visual SLAM (VSLAM) approach without reconstruction, mainly based on approaches for two-view loop closures. By restricting the map to a sparse keyframed pose graph without dense geometry representations, our `2GO' system achieves efficient optimization with competitive absolute trajectory accuracy. In particular, we find that recent advancements in image matching and monocular depth priors enable very accurate trajectory optimization without BA. We conduct extensive experiments on diverse datasets, including large-scale scenarios, and provide a detailed analysis of the trade-offs between runtime, accuracy, and map size. Our results demonstrate that this streamlined approach supports real-time performance, scales well in map size and trajectory duration, and effectively broadens the capabilities of VSLAM for long-duration deployments to large environments. |
|
| Deep Learning-Powered Visual SLAM Aimed at Assisting Visually Impaired Navigation | 2025-10-23 | ShowDespite advancements in SLAM technologies, robust operation under challenging conditions such as low-texture, motion-blur, or challenging lighting remains an open challenge. Such conditions are common in applications such as assistive navigation for the visually impaired. These challenges undermine localization accuracy and tracking stability, reducing navigation reliability and safety. To overcome these limitations, we present SELM-SLAM3, a deep learning-enhanced visual SLAM framework that integrates SuperPoint and LightGlue for robust feature extraction and matching. We evaluated our framework using TUM RGB-D, ICL-NUIM, and TartanAir datasets, which feature diverse and challenging scenarios. SELM-SLAM3 outperforms conventional ORB-SLAM3 by an average of 87.84% and exceeds state-of-the-art RGB-D SLAM systems by 36.77%. Our framework demonstrates enhanced performance under challenging conditions, such as low-texture scenes and fast motion, providing a reliable platform for developing navigation aids for the visually impaired. |
8 pag...8 pages, 7 figures, 4 tables |
| VAR-SLAM: Visual Adaptive and Robust SLAM for Dynamic Environments | 2025-10-17 | ShowVisual SLAM in dynamic environments remains challenging, as several existing methods rely on semantic filtering that only handles known object classes, or use fixed robust kernels that cannot adapt to unknown moving objects, leading to degraded accuracy when they appear in the scene. We present VAR-SLAM (Visual Adaptive and Robust SLAM), an ORB-SLAM3-based system that combines a lightweight semantic keypoint filter to deal with known moving objects, with Barron's adaptive robust loss to handle unknown ones. The shape parameter of the robust kernel is estimated online from residuals, allowing the system to automatically adjust between Gaussian and heavy-tailed behavior. We evaluate VAR-SLAM on the TUM RGB-D, Bonn RGB-D Dynamic, and OpenLORIS datasets, which include both known and unknown moving objects. Results show improved trajectory accuracy and robustness over state-of-the-art baselines, achieving up to 25% lower ATE RMSE than NGD-SLAM on challenging sequences, while maintaining performance at 27 FPS on average. |
Code ...Code available at https://github.com/iit-DLSLab/VAR-SLAM |
| Accelerated Feature Detectors for Visual SLAM: A Comparative Study of FPGA vs GPU | 2025-10-15 | ShowFeature detection is a common yet time-consuming module in Simultaneous Localization and Mapping (SLAM) implementations, which are increasingly deployed on power-constrained platforms, such as drones. Graphics Processing Units (GPUs) have been a popular accelerator for computer vision in general, and feature detection and SLAM in particular. On the other hand, System-on-Chips (SoCs) with integrated Field Programmable Gate Array (FPGA) are also widely available. This paper presents the first study of hardware-accelerated feature detectors considering a Visual SLAM (V-SLAM) pipeline. We offer new insights by comparing the best GPU-accelerated FAST, Harris, and SuperPoint implementations against the FPGA-accelerated counterparts on modern SoCs (Nvidia Jetson Orin and AMD Versal). The evaluation shows that when using a non-learning-based feature detector such as FAST and Harris, their GPU implementations, and the GPU-accelerated V-SLAM can achieve better run-time performance and energy efficiency than the FAST and Harris FPGA implementations as well as the FPGA-accelerated V-SLAM. However, when considering a learning-based detector such as SuperPoint, its FPGA implementation can achieve better run-time performance and energy efficiency (up to 3.1$\times$ and 1.4$\times$ improvements, respectively) than the GPU implementation. The FPGA-accelerated V-SLAM can also achieve comparable run-time performance compared to the GPU-accelerated V-SLAM, with better FPS in 2 out of 5 dataset sequences. When considering the accuracy, the results show that the GPU-accelerated V-SLAM is more accurate than the FPGA-accelerated V-SLAM in general. Last but not least, the use of hardware acceleration for feature detection could further improve the performance of the V-SLAM pipeline by having the global bundle adjustment module invoked less frequently without sacrificing accuracy. |
12 pages, 7 figures |
| SMapper: A Multi-Modal Data Acquisition Platform for SLAM Benchmarking | 2025-10-10 | ShowAdvancing research in fields such as Simultaneous Localization and Mapping (SLAM) and autonomous navigation critically depends on the availability of reliable and reproducible multimodal datasets. While several influential datasets have driven progress in these domains, they often suffer from limitations in sensing modalities, environmental diversity, and the reproducibility of the underlying hardware setups. To address these challenges, this paper introduces SMapper, a novel open-hardware, multi-sensor platform designed explicitly for, though not limited to, SLAM research. The device integrates synchronized LiDAR, multi-camera, and inertial sensing, supported by a robust calibration and synchronization pipeline that ensures precise spatio-temporal alignment across modalities. Its open and replicable design allows researchers to extend its capabilities and reproduce experiments across both handheld and robot-mounted scenarios. To demonstrate its practicality, we additionally release SMapper-light, a publicly available SLAM dataset containing representative indoor and outdoor sequences. The dataset includes tightly synchronized multimodal data and ground truth trajectories derived from offline LiDAR-based SLAM with sub-centimeter accuracy, alongside dense 3D reconstructions. Furthermore, the paper contains benchmarking results on state-of-the-art LiDAR and visual SLAM frameworks using the SMapper-light dataset. By combining open-hardware design, reproducible data collection, and comprehensive benchmarking, SMapper establishes a robust foundation for advancing SLAM algorithm development, evaluation, and reproducibility. The project's documentation, including source code, CAD models, and dataset links, is publicly available at https://snt-arg.github.io/smapper_docs. |
13 pa...13 pages, 5 figures, 6 tables |
| EgoExo++: Integrating On-demand Exocentric Visuals with 2.5D Ground Surface Estimation for Interactive Teleoperation of Subsea ROVs | 2025-10-08 | ShowUnderwater ROVs (Remotely Operated Vehicles) are indispensable for subsea exploration and task execution, yet typical teleoperation engines based on egocentric (first-person) video feeds restrict human operators' field-of-view and limit precise maneuvering in complex, unstructured underwater environments. To address this, we propose EgoExo, a geometry-driven solution integrated into a visual SLAM pipeline that synthesizes on-demand exocentric (third-person) views from egocentric camera feeds. Our proposed framework, EgoExo++, extends beyond 2D exocentric view synthesis (EgoExo) to augment a dense 2.5D ground surface estimation on-the-fly. It simultaneously renders the ROV model onto this reconstructed surface, enhancing semantic perception and depth comprehension. The computations involved are closed-form and rely solely on egocentric views and monocular SLAM estimates, which makes it portable across existing teleoperation engines and robust to varying waterbody characteristics. We validate the geometric accuracy of our approach through extensive experiments of 2-DOF indoor navigation and 6-DOF underwater cave exploration in challenging low-light conditions. Quantitative metrics confirm the reliability of the rendered Exo views, while a user study involving 15 operators demonstrates improved situational awareness, navigation safety, and task efficiency during teleoperation. Furthermore, we highlight the role of EgoExo++ augmented visuals in supporting shared autonomy, operator training, and embodied teleoperation. This new interactive approach to ROV teleoperation presents promising opportunities for future research in subsea telerobotics. |
EgoEx...EgoExo++ (Journal extension), V5, metadata updated, 12 pages |
| BIM Informed Visual SLAM for Construction Monitoring | 2025-10-08 | ShowSimultaneous Localization and Mapping (SLAM) is a key tool for monitoring construction sites, where aligning the evolving as-built state with the as-planned design enables early error detection and reduces costly rework. LiDAR-based SLAM achieves high geometric precision, but its sensors are typically large and power-demanding, limiting their use on portable platforms. Visual SLAM offers a practical alternative with lightweight cameras already embedded in most mobile devices. however, visually mapping construction environments remains challenging: repetitive layouts, occlusions, and incomplete or low-texture structures often cause drift in the trajectory map. To mitigate this, we propose an RGB-D SLAM system that incorporates the Building Information Model (BIM) as structural prior knowledge. Instead of relying solely on visual cues, our system continuously establishes correspondences between detected wall and their BIM counterparts, which are then introduced as constraints in the back-end optimization. The proposed method operates in real time and has been validated on real construction sites, reducing trajectory error by an average of 23.71% and map RMSE by 7.14% compared to visual SLAM baselines. These results demonstrate that BIM constraints enable reliable alignment of the digital plan with the as-built scene, even under partially constructed conditions. |
8 pag...8 pages, 5 tables, 4 figures |
| RSV-SLAM: Toward Real-Time Semantic Visual SLAM in Indoor Dynamic Environments | 2025-10-02 | ShowSimultaneous Localization and Mapping (SLAM) plays an important role in many robotics fields, including social robots. Many of the available visual SLAM methods are based on the assumption of a static world and struggle in dynamic environments. In the current study, we introduce a real-time semantic RGBD SLAM approach designed specifically for dynamic environments. Our proposed system can effectively detect moving objects and maintain a static map to ensure robust camera tracking. The key innovation of our approach is the incorporation of deep learning-based semantic information into SLAM systems to mitigate the impact of dynamic objects. Additionally, we enhance the semantic segmentation process by integrating an Extended Kalman filter to identify dynamic objects that may be temporarily idle. We have also implemented a generative network to fill in the missing regions of input images belonging to dynamic objects. This highly modular framework has been implemented on the ROS platform and can achieve around 22 fps on a GTX1080. Benchmarking the developed pipeline on dynamic sequences from the TUM dataset suggests that the proposed approach delivers competitive localization error in comparison with the state-of-the-art methods, all while operating in near real-time. The source code is publicly available. |
Proce...Proceedings of SAI Intelligent Systems Conference 2023 |
| Instant4D: 4D Gaussian Splatting in Minutes | 2025-10-01 | ShowDynamic view synthesis has seen significant advances, yet reconstructing scenes from uncalibrated, casual video remains challenging due to slow optimization and complex parameter estimation. In this work, we present Instant4D, a monocular reconstruction system that leverages native 4D representation to efficiently process casual video sequences within minutes, without calibrated cameras or depth sensors. Our method begins with geometric recovery through deep visual SLAM, followed by grid pruning to optimize scene representation. Our design significantly reduces redundancy while maintaining geometric integrity, cutting model size to under 10% of its original footprint. To handle temporal dynamics efficiently, we introduce a streamlined 4D Gaussian representation, achieving a 30x speed-up and reducing training time to within two minutes, while maintaining competitive performance across several benchmarks. Our method reconstruct a single video within 10 minutes on the Dycheck dataset or for a typical 200-frame video. We further apply our model to in-the-wild videos, showcasing its generalizability. Our project website is published at https://instant4d.github.io/. |
Accep...Accepted by NeurIPS 25 |
| Semantic Visual Simultaneous Localization and Mapping: A Survey on State of the Art, Challenges, and Future Directions | 2025-10-01 | ShowSemantic Simultaneous Localization and Mapping (SLAM) is a critical area of research within robotics and computer vision, focusing on the simultaneous localization of robotic systems and associating semantic information to construct the most accurate and complete comprehensive model of the surrounding environment. Since the first foundational work in Semantic SLAM appeared more than two decades ago, this field has received increasing attention across various scientific communities. Despite its significance, the field lacks comprehensive surveys encompassing recent advances and persistent challenges. In response, this study provides a thorough examination of the state-of-the-art of Semantic SLAM techniques, with the aim of illuminating current trends and key obstacles. Beginning with an in-depth exploration of the evolution of visual SLAM, this study outlines its strengths and unique characteristics, while also critically assessing previous survey literature. Subsequently, a unified problem formulation and evaluation of the modular solution framework is proposed, which divides the problem into discrete stages, including visual localization, semantic feature extraction, mapping, data association, and loop closure optimization. Moreover, this study investigates alternative methodologies such as deep learning and the utilization of large language models, alongside a review of relevant research about contemporary SLAM datasets. Concluding with a discussion on potential future research directions, this study serves as a comprehensive resource for researchers seeking to navigate the complex landscape of Semantic SLAM. |
|
| SuperEvent: Cross-Modal Learning of Event-based Keypoint Detection for SLAM | 2025-09-29 | ShowEvent-based keypoint detection and matching holds significant potential, enabling the integration of event sensors into highly optimized Visual SLAM systems developed for frame cameras over decades of research. Unfortunately, existing approaches struggle with the motion-dependent appearance of keypoints and the complex noise prevalent in event streams, resulting in severely limited feature matching capabilities and poor performance on downstream tasks. To mitigate this problem, we propose SuperEvent, a data-driven approach to predict stable keypoints with expressive descriptors. Due to the absence of event datasets with ground truth keypoint labels, we leverage existing frame-based keypoint detectors on readily available event-aligned and synchronized gray-scale frames for self-supervision: we generate temporally sparse keypoint pseudo-labels considering that events are a product of both scene appearance and camera motion. Combined with our novel, information-rich event representation, we enable SuperEvent to effectively learn robust keypoint detection and description in event streams. Finally, we demonstrate the usefulness of SuperEvent by its integration into a modern sparse keypoint and descriptor-based SLAM framework originally developed for traditional cameras, surpassing the state-of-the-art in event-based SLAM by a wide margin. Source code is available at https://ethz-mrl.github.io/SuperEvent/. |
|
| GRS-SLAM3R: Real-Time Dense SLAM with Gated Recurrent State | 2025-09-28 | ShowDUSt3R-based end-to-end scene reconstruction has recently shown promising results in dense visual SLAM. However, most existing methods only use image pairs to estimate pointmaps, overlooking spatial memory and global consistency.To this end, we introduce GRS-SLAM3R, an end-to-end SLAM framework for dense scene reconstruction and pose estimation from RGB images without any prior knowledge of the scene or camera parameters. Unlike existing DUSt3R-based frameworks, which operate on all image pairs and predict per-pair point maps in local coordinate frames, our method supports sequentialized input and incrementally estimates metric-scale point clouds in the global coordinate. In order to improve consistent spatial correlation, we use a latent state for spatial memory and design a transformer-based gated update module to reset and update the spatial memory that continuously aggregates and tracks relevant 3D information across frames. Furthermore, we partition the scene into submaps, apply local alignment within each submap, and register all submaps into a common world frame using relative constraints, producing a globally consistent map. Experiments on various datasets show that our framework achieves superior reconstruction accuracy while maintaining real-time performance. |
|
| Good Weights: Proactive, Adaptive Dead Reckoning Fusion for Continuous and Robust Visual SLAM | 2025-09-26 | ShowGiven that Visual SLAM relies on appearance cues for localization and scene understanding, texture-less or visually degraded environments (e.g., plain walls or low lighting) lead to poor pose estimation and track loss. However, robots are typically equipped with sensors that provide some form of dead reckoning odometry with reasonable short-time performance but unreliable long-time performance. The Good Weights (GW) algorithm described here provides a framework to adaptively integrate dead reckoning (DR) with passive visual SLAM for continuous and accurate frame-level pose estimation. Importantly, it describes how all modules in a comprehensive SLAM system must be modified to incorporate DR into its design. Adaptive weighting increases DR influence when visual tracking is unreliable and reduces when visual feature information is strong, maintaining pose track without overreliance on DR. Good Weights yields a practical solution for mobile navigation that improves visual SLAM performance and robustness. Experiments on collected datasets and in real-world deployment demonstrate the benefits of Good Weights. |
8 pag...8 pages, 9 figures, 1 table. Submitted to IEEE Conference |
| Optical Ocean Recipes: Creating Realistic Datasets to Facilitate Underwater Vision Research | 2025-09-24 | ShowThe development and evaluation of machine vision in underwater environments remains challenging, often relying on trial-and-error-based testing tailored to specific applications. This is partly due to the lack of controlled, ground-truthed testing environments that account for the optical challenges, such as color distortion from spectrally variant light attenuation, reduced contrast and blur from backscatter and volume scattering, and dynamic light patterns from natural or artificial illumination. Additionally, the appearance of ocean water in images varies significantly across regions, depths, and seasons. However, most machine vision evaluations are conducted under specific optical water types and imaging conditions, therefore often lack generalizability. Exhaustive testing across diverse open-water scenarios is technically impractical. To address this, we introduce the \textit{Optical Ocean Recipes}, a framework for creating realistic datasets under controlled underwater conditions. Unlike synthetic or open-water data, these recipes, using calibrated color and scattering additives, enable repeatable and controlled testing of the impact of water composition on image appearance. Hence, this provides a unique framework for analyzing machine vision in realistic, yet controlled underwater scenarios. The controlled environment enables the creation of ground-truth data for a range of vision tasks, including water parameter estimation, image restoration, segmentation, visual SLAM, and underwater image synthesis. We provide a demonstration dataset generated using the Optical Ocean Recipes and briefly demonstrate the use of our system for two underwater vision tasks. The dataset and evaluation code will be made available. |
26 pa...26 pages, 9 figures, submitted to IEEE Journal of Ocean Engineering |
| ConfidentSplat: Confidence-Weighted Depth Fusion for Accurate 3D Gaussian Splatting SLAM | 2025-09-21 | ShowWe introduce ConfidentSplat, a novel 3D Gaussian Splatting (3DGS)-based SLAM system for robust, highfidelity RGB-only reconstruction. Addressing geometric inaccuracies in existing RGB-only 3DGS SLAM methods that stem from unreliable depth estimation, ConfidentSplat incorporates a core innovation: a confidence-weighted fusion mechanism. This mechanism adaptively integrates depth cues from multiview geometry with learned monocular priors (Omnidata ViT), dynamically weighting their contributions based on explicit reliability estimates-derived predominantly from multi-view geometric consistency-to generate high-fidelity proxy depth for map supervision. The resulting proxy depth guides the optimization of a deformable 3DGS map, which efficiently adapts online to maintain global consistency following pose updates from a DROID-SLAM-inspired frontend and backend optimizations (loop closure, global bundle adjustment). Extensive validation on standard benchmarks (TUM-RGBD, ScanNet) and diverse custom mobile datasets demonstrates significant improvements in reconstruction accuracy (L1 depth error) and novel view synthesis fidelity (PSNR, SSIM, LPIPS) over baselines, particularly in challenging conditions. ConfidentSplat underscores the efficacy of principled, confidence-aware sensor fusion for advancing state-of-the-art dense visual SLAM. |
|
| FastTrack: GPU-Accelerated Tracking for Visual SLAM | 2025-09-13 | ShowThe tracking module of a visual-inertial SLAM system processes incoming image frames and IMU data to estimate the position of the frame in relation to the map. It is important for the tracking to complete in a timely manner for each frame to avoid poor localization or tracking loss. We therefore present a new approach which leverages GPU computing power to accelerate time-consuming components of tracking in order to improve its performance. These components include stereo feature matching and local map tracking. We implement our design inside the ORB-SLAM3 tracking process using CUDA. Our evaluation demonstrates an overall improvement in tracking performance of up to 2.8x on a desktop and Jetson Xavier NX board in stereo-inertial mode, using the well-known SLAM datasets EuRoC and TUM-VI. |
Accep...Accepted for presentation at IROS 2025, preprint |
| PINGS: Gaussian Splatting Meets Distance Fields within a Point-Based Implicit Neural Map | 2025-09-09 | ShowRobots benefit from high-fidelity reconstructions of their environment, which should be geometrically accurate and photorealistic to support downstream tasks. While this can be achieved by building distance fields from range sensors and radiance fields from cameras, realising scalable incremental mapping of both fields consistently and at the same time with high quality is challenging. In this paper, we propose a novel map representation that unifies a continuous signed distance field and a Gaussian splatting radiance field within an elastic and compact point-based implicit neural map. By enforcing geometric consistency between these fields, we achieve mutual improvements by exploiting both modalities. We present a novel LiDAR-visual SLAM system called PINGS using the proposed map representation and evaluate it on several challenging large-scale datasets. Experimental results demonstrate that PINGS can incrementally build globally consistent distance and radiance fields encoded with a compact set of neural points. Compared to state-of-the-art methods, PINGS achieves superior photometric and geometric rendering at novel views by constraining the radiance field with the distance field. Furthermore, by utilizing dense photometric cues and multi-view consistency from the radiance field, PINGS produces more accurate distance fields, leading to improved odometry estimation and mesh reconstruction. We also provide an open-source implementation of PING at: https://github.com/PRBonn/PINGS. |
15 pa...15 pages, 8 figures, presented at RSS 2025 |
| Active Illumination for Visual Ego-Motion Estimation in the Dark | 2025-09-08 | ShowVisual Odometry (VO) and Visual SLAM (V-SLAM) systems often struggle in low-light and dark environments due to the lack of robust visual features. In this paper, we propose a novel active illumination framework to enhance the performance of VO and V-SLAM algorithms in these challenging conditions. The developed approach dynamically controls a moving light source to illuminate highly textured areas, thereby improving feature extraction and tracking. Specifically, a detector block, which incorporates a deep learning-based enhancing network, identifies regions with relevant features. Then, a pan-tilt controller is responsible for guiding the light beam toward these areas, so that to provide information-rich images to the ego-motion estimation algorithm. Experimental results on a real robotic platform demonstrate the effectiveness of the proposed method, showing a reduction in the pose estimation error up to 75% with respect to a traditional fixed lighting technique. |
|
| FGO-SLAM: Enhancing Gaussian SLAM with Globally Consistent Opacity Radiance Field | 2025-09-01 | ShowVisual SLAM has regained attention due to its ability to provide perceptual capabilities and simulation test data for Embodied AI. However, traditional SLAM methods struggle to meet the demands of high-quality scene reconstruction, and Gaussian SLAM systems, despite their rapid rendering and high-quality mapping capabilities, lack effective pose optimization methods and face challenges in geometric reconstruction. To address these issues, we introduce FGO-SLAM, a Gaussian SLAM system that employs an opacity radiance field as the scene representation to enhance geometric mapping performance. After initial pose estimation, we apply global adjustment to optimize camera poses and sparse point cloud, ensuring robust tracking of our approach. Additionally, we maintain a globally consistent opacity radiance field based on 3D Gaussians and introduce depth distortion and normal consistency terms to refine the scene representation. Furthermore, after constructing tetrahedral grids, we identify level sets to directly extract surfaces from 3D Gaussians. Results across various real-world and large-scale synthetic datasets demonstrate that our method achieves state-of-the-art tracking accuracy and mapping performance. |
ICRA 2025 |
| DyPho-SLAM : Real-time Photorealistic SLAM in Dynamic Environments | 2025-08-31 | ShowVisual SLAM algorithms have been enhanced through the exploration of Gaussian Splatting representations, particularly in generating high-fidelity dense maps. While existing methods perform reliably in static environments, they often encounter camera tracking drift and fuzzy mapping when dealing with the disturbances caused by moving objects. This paper presents DyPho-SLAM, a real-time, resource-efficient visual SLAM system designed to address the challenges of localization and photorealistic mapping in environments with dynamic objects. Specifically, the proposed system integrates prior image information to generate refined masks, effectively minimizing noise from mask misjudgment. Additionally, to enhance constraints for optimization after removing dynamic obstacles, we devise adaptive feature extraction strategies significantly improving the system's resilience. Experiments conducted on publicly dynamic RGB-D datasets demonstrate that the proposed system achieves state-of-the-art performance in camera pose estimation and dense map reconstruction, while operating in real-time in dynamic scenes. |
Accep...Accepted by ICME 2025(Oral) |
| Survey on Monocular Metric Depth Estimation | 2025-08-26 | ShowMonocular Depth Estimation (MDE) enables spatial understanding, 3D reconstruction, and autonomous navigation, yet deep learning approaches often predict only relative depth without a consistent metric scale. This limitation reduces reliability in applications such as visual SLAM, precise 3D modeling, and view synthesis. Monocular Metric Depth Estimation (MMDE) overcomes this challenge by producing depth maps with absolute scale, ensuring geometric consistency and enabling deployment without additional calibration. This survey reviews the evolution of MMDE, from geometry-based methods to state-of-the-art deep models, with emphasis on the datasets that drive progress. Key benchmarks, including KITTI, NYU-D, ApolloScape, and TartanAir, are examined in terms of modality, scene type, and application domain. Methodological advances are analyzed, covering domain generalization, boundary preservation, and the integration of synthetic and real data. Techniques such as unsupervised and semi-supervised learning, patch-based inference, architectural innovations, and generative modeling are evaluated for their strengths and limitations. By synthesizing current progress, highlighting the importance of high-quality datasets, and identifying open challenges, this survey provides a structured reference for advancing MMDE and supporting its adoption in real-world computer vision systems. |
|
| MCN-SLAM: Multi-Agent Collaborative Neural SLAM with Hybrid Implicit Neural Scene Representation | 2025-08-19 | ShowNeural implicit scene representations have recently shown promising results in dense visual SLAM. However, existing implicit SLAM algorithms are constrained to single-agent scenarios, and fall difficulties in large-scale scenes and long sequences. Existing NeRF-based multi-agent SLAM frameworks cannot meet the constraints of communication bandwidth. To this end, we propose the first distributed multi-agent collaborative neural SLAM framework with hybrid scene representation, distributed camera tracking, intra-to-inter loop closure, and online distillation for multiple submap fusion. A novel triplane-grid joint scene representation method is proposed to improve scene reconstruction. A novel intra-to-inter loop closure method is designed to achieve local (single-agent) and global (multi-agent) consistency. We also design a novel online distillation method to fuse the information of different submaps to achieve global consistency. Furthermore, to the best of our knowledge, there is no real-world dataset for NeRF-based/GS-based SLAM that provides both continuous-time trajectories groundtruth and high-accuracy 3D meshes groundtruth. To this end, we propose the first real-world Dense slam (DES) dataset covering both single-agent and multi-agent scenarios, ranging from small rooms to large-scale outdoor scenes, with high-accuracy ground truth for both 3D mesh and continuous-time camera trajectory. This dataset can advance the development of the research in both SLAM, 3D reconstruction, and visual foundation model. Experiments on various datasets demonstrate the superiority of the proposed method in both mapping, tracking, and communication. The dataset and code will open-source on https://github.com/dtc111111/mcnslam. |
|
| Pseudo Depth Meets Gaussian: A Feed-forward RGB SLAM Baseline | 2025-08-06 | ShowIncrementally recovering real-sized 3D geometry from a pose-free RGB stream is a challenging task in 3D reconstruction, requiring minimal assumptions on input data. Existing methods can be broadly categorized into end-to-end and visual SLAM-based approaches, both of which either struggle with long sequences or depend on slow test-time optimization and depth sensors. To address this, we first integrate a depth estimator into an RGB-D SLAM system, but this approach is hindered by inaccurate geometric details in predicted depth. Through further investigation, we find that 3D Gaussian mapping can effectively solve this problem. Building on this, we propose an online 3D reconstruction method using 3D Gaussian-based SLAM, combined with a feed-forward recurrent prediction module to directly infer camera pose from optical flow. This approach replaces slow test-time optimization with fast network inference, significantly improving tracking speed. Additionally, we introduce a local graph rendering technique to enhance robustness in feed-forward pose prediction. Experimental results on the Replica and TUM-RGBD datasets, along with a real-world deployment demonstration, show that our method achieves performance on par with the state-of-the-art SplaTAM, while reducing tracking time by more than 90%. |
IROS 2025 |
| pySLAM: An Open-Source, Modular, and Extensible Framework for SLAM | 2025-08-02 | ShowpySLAM is an open-source Python framework for Visual SLAM that supports monocular, stereo, and RGB-D camera inputs. It offers a flexible and modular interface, integrating a broad range of both classical and learning-based local features. The framework includes multiple loop closure strategies, a volumetric reconstruction pipeline, and support for depth prediction models. It also offers a comprehensive set of tools for experimenting with and evaluating visual odometry and SLAM modules. Designed for both beginners and experienced researchers, pySLAM emphasizes rapid prototyping, extensibility, and reproducibility across diverse datasets. Its modular architecture facilitates the integration of custom components and encourages research that bridges traditional and deep learning-based approaches. Community contributions are welcome, fostering collaborative development and innovation in the field of Visual SLAM. This document presents the pySLAM framework, outlining its main components, features, and usage. |
|
| Lost in Tracking Translation: A Comprehensive Analysis of Visual SLAM in Human-Centered XR and IoT Ecosystems | 2025-07-17 | ShowAdvancements in tracking algorithms have empowered nascent applications across various domains, from steering autonomous vehicles to guiding robots to enhancing augmented reality experiences for users. However, these algorithms are application-specific and do not work across applications with different types of motion; even a tracking algorithm designed for a given application does not work in scenarios deviating from highly standard conditions. For example, a tracking algorithm designed for robot navigation inside a building will not work for tracking the same robot in an outdoor environment. To demonstrate this problem, we evaluate the performance of the state-of-the-art tracking methods across various applications and scenarios. To inform our analysis, we first categorize algorithmic, environmental, and locomotion-related challenges faced by tracking algorithms. We quantitatively evaluate the performance using multiple tracking algorithms and representative datasets for a wide range of Internet of Things (IoT) and Extended Reality (XR) applications, including autonomous vehicles, drones, and humans. Our analysis shows that no tracking algorithm works across different applications and scenarios within applications. Ultimately, using the insights generated from our analysis, we discuss multiple approaches to improving the tracking performance using input data characterization, leveraging intermediate information, and output evaluation. |
|
| DINO-VO: A Feature-based Visual Odometry Leveraging a Visual Foundation Model | 2025-07-17 | ShowLearning-based monocular visual odometry (VO) poses robustness, generalization, and efficiency challenges in robotics. Recent advances in visual foundation models, such as DINOv2, have improved robustness and generalization in various vision tasks, yet their integration in VO remains limited due to coarse feature granularity. In this paper, we present DINO-VO, a feature-based VO system leveraging DINOv2 visual foundation model for its sparse feature matching. To address the integration challenge, we propose a salient keypoints detector tailored to DINOv2's coarse features. Furthermore, we complement DINOv2's robust-semantic features with fine-grained geometric features, resulting in more localizable representations. Finally, a transformer-based matcher and differentiable pose estimation layer enable precise camera motion estimation by learning good matches. Against prior detector-descriptor networks like SuperPoint, DINO-VO demonstrates greater robustness in challenging environments. Furthermore, we show superior accuracy and generalization of the proposed feature descriptors against standalone DINOv2 coarse features. DINO-VO outperforms prior frame-to-frame VO methods on the TartanAir and KITTI datasets and is competitive on EuRoC dataset, while running efficiently at 72 FPS with less than 1GB of memory usage on a single GPU. Moreover, it performs competitively against Visual SLAM systems on outdoor driving scenarios, showcasing its generalization capabilities. |
8 pag...8 pages, 6 figures. Accepted for publication in IEEE Robotics and Automation Letters (RA-L), July 2025 |
| FLAF: Focal Line and Feature-constrained Active View Planning for Visual Teach and Repeat | 2025-07-15 | ShowThis paper presents FLAF, a focal line and feature-constrained active view planning method for tracking failure avoidance in feature-based visual navigation of mobile robots. Our FLAF-based visual navigation is built upon a feature-based visual teach and repeat (VT&R) framework, which supports many robotic applications by teaching a robot to navigate on various paths that cover a significant portion of daily autonomous navigation requirements. However, tracking failure in feature-based visual simultaneous localization and mapping (VSLAM) caused by textureless regions in human-made environments is still limiting VT&R to be adopted in the real world. To address this problem, the proposed view planner is integrated into a feature-based visual SLAM system to build up an active VT&R system that avoids tracking failure. In our system, a pan-tilt unit (PTU)-based active camera is mounted on the mobile robot. Using FLAF, the active camera-based VSLAM operates during the teaching phase to construct a complete path map and in the repeat phase to maintain stable localization. FLAF orients the robot toward more map points to avoid mapping failures during path learning and toward more feature-identifiable map points beneficial for localization while following the learned trajectory. Experiments in real scenarios demonstrate that FLAF outperforms the methods that do not consider feature-identifiability, and our active VT&R system performs well in complex environments by effectively dealing with low-texture regions. |
|
| IRAF-SLAM: An Illumination-Robust and Adaptive Feature-Culling Front-End for Visual SLAM in Challenging Environments | 2025-07-10 | ShowRobust Visual SLAM (vSLAM) is essential for autonomous systems operating in real-world environments, where challenges such as dynamic objects, low texture, and critically, varying illumination conditions often degrade performance. Existing feature-based SLAM systems rely on fixed front-end parameters, making them vulnerable to sudden lighting changes and unstable feature tracking. To address these challenges, we propose ``IRAF-SLAM'', an Illumination-Robust and Adaptive Feature-Culling front-end designed to enhance vSLAM resilience in complex and challenging environments. Our approach introduces: (1) an image enhancement scheme to preprocess and adjust image quality under varying lighting conditions; (2) an adaptive feature extraction mechanism that dynamically adjusts detection sensitivity based on image entropy, pixel intensity, and gradient analysis; and (3) a feature culling strategy that filters out unreliable feature points using density distribution analysis and a lighting impact factor. Comprehensive evaluations on the TUM-VI and European Robotics Challenge (EuRoC) datasets demonstrate that IRAF-SLAM significantly reduces tracking failures and achieves superior trajectory accuracy compared to state-of-the-art vSLAM methods under adverse illumination conditions. These results highlight the effectiveness of adaptive front-end strategies in improving vSLAM robustness without incurring significant computational overhead. The implementation of IRAF-SLAM is publicly available at https://thanhnguyencanh. github.io/IRAF-SLAM/. |
In th...In the European Conference on Mobile Robots 2025 |
| ROVER: A Multi-Season Dataset for Visual SLAM | 2025-07-09 | ShowRobust SLAM is a crucial enabler for autonomous navigation in natural, semi-structured environments such as parks and gardens. However, these environments present unique challenges for SLAM due to frequent seasonal changes, varying light conditions, and dense vegetation. These factors often degrade the performance of visual SLAM algorithms originally developed for structured urban environments. To address this gap, we present ROVER, a comprehensive benchmark dataset tailored for evaluating visual SLAM algorithms under diverse environmental conditions and spatial configurations. We captured the dataset with a robotic platform equipped with monocular, stereo, and RGBD cameras, as well as inertial sensors. It covers 39 recordings across five outdoor locations, collected through all seasons and various lighting scenarios, i.e., day, dusk, and night with and without external lighting. With this novel dataset, we evaluate several traditional and deep learning-based SLAM methods and study their performance in diverse challenging conditions. The results demonstrate that while stereo-inertial and RGBD configurations generally perform better under favorable lighting and moderate vegetation, most SLAM systems perform poorly in low-light and high-vegetation scenarios, particularly during summer and autumn. Our analysis highlights the need for improved adaptability in visual SLAM algorithms for outdoor applications, as current systems struggle with dynamic environmental factors affecting scale, feature extraction, and trajectory consistency. This dataset provides a solid foundation for advancing visual SLAM research in real-world, semi-structured environments, fostering the development of more resilient SLAM systems for long-term outdoor localization and mapping. The dataset and the code of the benchmark are available under https://iis-esslingen.github.io/rover. |
Copyr...Copyright 2025 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works |
| Real-Time Obstacle Avoidance Algorithms for Unmanned Aerial and Ground Vehicles | 2025-06-25 | ShowThe growing use of mobile robots in sectors such as automotive, agriculture, and rescue operations reflects progress in robotics and autonomy. In unmanned aerial vehicles (UAVs), most research emphasizes visual SLAM, sensor fusion, and path planning. However, applying UAVs to search and rescue missions in disaster zones remains underexplored, especially for autonomous navigation. This report develops methods for real-time and secure UAV maneuvering in complex 3D environments, crucial during forest fires. Building upon past research, it focuses on designing navigation algorithms for unfamiliar and hazardous environments, aiming to improve rescue efficiency and safety through UAV-based early warning and rapid response. The work unfolds in phases. First, a 2D fusion navigation strategy is explored, initially for mobile robots, enabling safe movement in dynamic settings. This sets the stage for advanced features such as adaptive obstacle handling and decision-making enhancements. Next, a novel 3D reactive navigation strategy is introduced for collision-free movement in forest fire simulations, addressing the unique challenges of UAV operations in such scenarios. Finally, the report proposes a unified control approach that integrates UAVs and unmanned ground vehicles (UGVs) for coordinated rescue missions in forest environments. Each phase presents challenges, proposes control models, and validates them with mathematical and simulation-based evidence. The study offers practical value and academic insights for improving the role of UAVs in natural disaster rescue operations. |
|
| Neural Graph Map: Dense Mapping with Efficient Loop Closure Integration | 2025-06-25 | ShowNeural field-based SLAM methods typically employ a single, monolithic field as their scene representation. This prevents efficient incorporation of loop closure constraints and limits scalability. To address these shortcomings, we propose a novel RGB-D neural mapping framework in which the scene is represented by a collection of lightweight neural fields which are dynamically anchored to the pose graph of a sparse visual SLAM system. Our approach shows the ability to integrate large-scale loop closures, while requiring only minimal reintegration. Furthermore, we verify the scalability of our approach by demonstrating successful building-scale mapping taking multiple loop closures into account during the optimization, and show that our method outperforms existing state-of-the-art approaches on large scenes in terms of quality and runtime. Our code is available open-source at https://github.com/KTH-RPL/neural_graph_mapping. |
WACV ...WACV 2025, Project page: https://kth-rpl.github.io/neural_graph_mapping/ |
| Multimodal Fusion SLAM with Fourier Attention | 2025-06-24 | ShowVisual SLAM is particularly challenging in environments affected by noise, varying lighting conditions, and darkness. Learning-based optical flow algorithms can leverage multiple modalities to address these challenges, but traditional optical flow-based visual SLAM approaches often require significant computational resources.To overcome this limitation, we propose FMF-SLAM, an efficient multimodal fusion SLAM method that utilizes fast Fourier transform (FFT) to enhance the algorithm efficiency. Specifically, we introduce a novel Fourier-based self-attention and cross-attention mechanism to extract features from RGB and depth signals. We further enhance the interaction of multimodal features by incorporating multi-scale knowledge distillation across modalities. We also demonstrate the practical feasibility of FMF-SLAM in real-world scenarios with real time performance by integrating it with a security robot by fusing with a global positioning module GNSS-RTK and global Bundle Adjustment. Our approach is validated using video sequences from TUM, TartanAir, and our real-world datasets, showcasing state-of-the-art performance under noisy, varying lighting, and dark conditions.Our code and datasets are available at https://github.com/youjie-zhou/FMF-SLAM.git. |
Accepted in IEEE RAL |
| GRAND-SLAM: Local Optimization for Globally Consistent Large-Scale Multi-Agent Gaussian SLAM | 2025-06-23 | Show3D Gaussian splatting has emerged as an expressive scene representation for RGB-D visual SLAM, but its application to large-scale, multi-agent outdoor environments remains unexplored. Multi-agent Gaussian SLAM is a promising approach to rapid exploration and reconstruction of environments, offering scalable environment representations, but existing approaches are limited to small-scale, indoor environments. To that end, we propose Gaussian Reconstruction via Multi-Agent Dense SLAM, or GRAND-SLAM, a collaborative Gaussian splatting SLAM method that integrates i) an implicit tracking module based on local optimization over submaps and ii) an approach to inter- and intra-robot loop closure integrated into a pose-graph optimization framework. Experiments show that GRAND-SLAM provides state-of-the-art tracking performance and 28% higher PSNR than existing methods on the Replica indoor dataset, as well as 91% lower multi-agent tracking error and improved rendering over existing multi-agent methods on the large-scale, outdoor Kimera-Multi dataset. |
|
| 4Seasons: Benchmarking Visual SLAM and Long-Term Localization for Autonomous Driving in Challenging Conditions | 2025-06-19 | ShowIn this paper, we present a novel visual SLAM and long-term localization benchmark for autonomous driving in challenging conditions based on the large-scale 4Seasons dataset. The proposed benchmark provides drastic appearance variations caused by seasonal changes and diverse weather and illumination conditions. While significant progress has been made in advancing visual SLAM on small-scale datasets with similar conditions, there is still a lack of unified benchmarks representative of real-world scenarios for autonomous driving. We introduce a new unified benchmark for jointly evaluating visual odometry, global place recognition, and map-based visual localization performance which is crucial to successfully enable autonomous driving in any condition. The data has been collected for more than one year, resulting in more than 300 km of recordings in nine different environments ranging from a multi-level parking garage to urban (including tunnels) to countryside and highway. We provide globally consistent reference poses with up to centimeter-level accuracy obtained from the fusion of direct stereo-inertial odometry with RTK GNSS. We evaluate the performance of several state-of-the-art visual odometry and visual localization baseline approaches on the benchmark and analyze their properties. The experimental results provide new insights into current approaches and show promising potential for future research. Our benchmark and evaluation protocols will be available at https://go.vision.in.tum.de/4seasons. |
Publi...Published in International Journal of Computer Vision (IJCV). arXiv admin note: substantial text overlap with arXiv:2009.06364 |
| NGD-SLAM: Towards Real-Time Dynamic SLAM without GPU | 2025-06-16 | ShowMany existing visual SLAM methods can achieve high localization accuracy in dynamic environments by leveraging deep learning to mask moving objects. However, these methods incur significant computational overhead as the camera tracking needs to wait for the deep neural network to generate mask at each frame, and they typically require GPUs for real-time operation, which restricts their practicality in real-world robotic applications. Therefore, this paper proposes a real-time dynamic SLAM system that runs exclusively on a CPU. Our approach incorporates a mask propagation mechanism that decouples camera tracking and deep learning-based masking for each frame. We also introduce a hybrid tracking strategy that integrates ORB features with optical flow methods, enhancing both robustness and efficiency by selectively allocating computational resources to input frames. Compared to previous methods, our system maintains high localization accuracy in dynamic environments while achieving a tracking frame rate of 60 FPS on a laptop CPU. These results demonstrate the feasibility of utilizing deep learning for dynamic SLAM without GPU support. Since most existing dynamic SLAM systems are not open-source, we make our code publicly available at: https://github.com/yuhaozhang7/NGD-SLAM |
7 pages, 6 figures |
| LRSLAM: Low-rank Representation of Signed Distance Fields in Dense Visual SLAM System | 2025-06-12 | ShowSimultaneous Localization and Mapping (SLAM) has been crucial across various domains, including autonomous driving, mobile robotics, and mixed reality. Dense visual SLAM, leveraging RGB-D camera systems, offers advantages but faces challenges in achieving real-time performance, robustness, and scalability for large-scale scenes. Recent approaches utilizing neural implicit scene representations show promise but suffer from high computational costs and memory requirements. ESLAM introduced a plane-based tensor decomposition but still struggled with memory growth. Addressing these challenges, we propose a more efficient visual SLAM model, called LRSLAM, utilizing low-rank tensor decomposition methods. Our approach, leveraging the Six-axis and CP decompositions, achieves better convergence rates, memory efficiency, and reconstruction/localization quality than existing state-of-the-art approaches. Evaluation across diverse indoor RGB-D datasets demonstrates LRSLAM's superior performance in terms of parameter efficiency, processing time, and accuracy, retaining reconstruction and localization quality. Our code will be publicly available upon publication. |
Accep...Accepted at ECCV 2024 |
| VAULT: A Mobile Mapping System for ROS 2-based Autonomous Robots | 2025-06-11 | ShowLocalization plays a crucial role in the navigation capabilities of autonomous robots, and while indoor environments can rely on wheel odometry and 2D LiDAR-based mapping, outdoor settings such as agriculture and forestry, present unique challenges that necessitate real-time localization and consistent mapping. Addressing this need, this paper introduces the VAULT prototype, a ROS 2-based mobile mapping system (MMS) that combines various sensors to enable robust outdoor and indoor localization. The proposed solution harnesses the power of Global Navigation Satellite System (GNSS) data, visual-inertial odometry (VIO), inertial measurement unit (IMU) data, and the Extended Kalman Filter (EKF) to generate reliable 3D odometry. To further enhance the localization accuracy, Visual SLAM (VSLAM) is employed, resulting in the creation of a comprehensive 3D point cloud map. By leveraging these sensor technologies and advanced algorithms, the prototype offers a comprehensive solution for outdoor localization in autonomous mobile robots, enabling them to navigate and map their surroundings with confidence and precision. |
15 pa...15 pages, 5 figures, Submitted to WAF 2023: Workshop de Agentes Fisicos |
| AquaticVision: Benchmarking Visual SLAM in Underwater Environment with Events and Frames | 2025-06-05 | ShowMany underwater applications, such as offshore asset inspections, rely on visual inspection and detailed 3D reconstruction. Recent advancements in underwater visual SLAM systems for aquatic environments have garnered significant attention in marine robotics research. However, existing underwater visual SLAM datasets often lack groundtruth trajectory data, making it difficult to objectively compare the performance of different SLAM algorithms based solely on qualitative results or COLMAP reconstruction. In this paper, we present a novel underwater dataset that includes ground truth trajectory data obtained using a motion capture system. Additionally, for the first time, we release visual data that includes both events and frames for benchmarking underwater visual positioning. By providing event camera data, we aim to facilitate the development of more robust and advanced underwater visual SLAM algorithms. The use of event cameras can help mitigate challenges posed by extremely low light or hazy underwater conditions. The webpage of our dataset is https://sites.google.com/view/aquaticvision-lias. |
|
| PLGSLAM: Progressive Neural Scene Represenation with Local to Global Bundle Adjustment | 2025-05-27 | ShowNeural implicit scene representations have recently shown encouraging results in dense visual SLAM. However, existing methods produce low-quality scene reconstruction and low-accuracy localization performance when scaling up to large indoor scenes and long sequences. These limitations are mainly due to their single, global radiance field with finite capacity, which does not adapt to large scenarios. Their end-to-end pose networks are also not robust enough with the growth of cumulative errors in large scenes. To this end, we introduce PLGSLAM, a neural visual SLAM system capable of high-fidelity surface reconstruction and robust camera tracking in real-time. To handle large-scale indoor scenes, PLGSLAM proposes a progressive scene representation method which dynamically allocates new local scene representation trained with frames within a local sliding window. This allows us to scale up to larger indoor scenes and improves robustness (even under pose drifts). In local scene representation, PLGSLAM utilizes tri-planes for local high-frequency features with multi-layer perceptron (MLP) networks for the low-frequency feature, achieving smoothness and scene completion in unobserved areas. Moreover, we propose local-to-global bundle adjustment method with a global keyframe database to address the increased pose drifts on long sequences. Experimental results demonstrate that PLGSLAM achieves state-of-the-art scene reconstruction results and tracking performance across various datasets and scenarios (both in small and large-scale indoor environments). The code is open-sourced at https://github.com/dtc111111/plgslam. |
Accep...Accepted by CVPR 2024 |
| TAT-VPR: Ternary Adaptive Transformer for Dynamic and Efficient Visual Place Recognition | 2025-05-22 | ShowTAT-VPR is a ternary-quantized transformer that brings dynamic accuracy-efficiency trade-offs to visual SLAM loop-closure. By fusing ternary weights with a learned activation-sparsity gate, the model can control computation by up to 40% at run-time without degrading performance (Recall@1). The proposed two-stage distillation pipeline preserves descriptor quality, letting it run on micro-UAV and embedded SLAM stacks while matching state-of-the-art localization accuracy. |
|
| Is Semantic SLAM Ready for Embedded Systems ? A Comparative Survey | 2025-0 |