Tada is a platform that allows us to quickly test interaction ideas on the following thesis: our future models will deeply know our contexts and who we are---so well that they'll be able to predict what we'll do next.
You might be here to test out our first interaction:
Visit the site: https://generalusermodels.github.io/tada/
Download the latest release here.
Make sure you have uv installed first. All you have to do then is:
git clone git@github.com:GeneralUserModels/tada.git
cd tada
npm install
npm run dev
Tada 🎉 builds on research across a few academic papers. If you're interested, you can read or cite the papers below!
Creating General User Models from Computer Use
@inproceedings{shaikh2025creating,
title={Creating general user models from computer use},
author={Shaikh, Omar and Sapkota, Shardul and Rizvi, Shan and Horvitz, Eric and Park, Joon Sung and Yang, Diyi and Bernstein, Michael S},
booktitle={Proceedings of the 38th Annual ACM Symposium on User Interface Software and Technology},
pages={1--23},
year={2025}
}Learning Next Action Predictors from Human-Computer Interaction
@article{shaikh2026learning,
title={Learning Next Action Predictors from Human-Computer Interaction},
author={Shaikh, Omar and Teutschbein, Valentin and Gandhi, Kanishk and Chi, Yikun and Haber, Nick and Robinson, Thomas and Ram, Nilam and Reeves, Byron and Yang, Sherry and Bernstein, Michael S and others},
journal={arXiv preprint arXiv:2603.05923},
year={2026}
}Just-In-Time Objectives: A General Approach for Specialized AI Interactions
@article{lam2025just,
title={Just-In-Time Objectives: A General Approach for Specialized AI Interactions},
author={Lam, Michelle S and Shaikh, Omar and Xu, Hallie and Guo, Alice and Yang, Diyi and Heer, Jeffrey and Landay, James A and Bernstein, Michael S},
journal={arXiv preprint arXiv:2510.14591},
year={2025}
}Learning to Simulate Human Dialogue
@article{gandhi2026learning,
title={Learning to Simulate Human Dialogue},
author={Gandhi, Kanishk and Bhatia, Agam and Goodman, Noah D},
journal={arXiv preprint arXiv:2601.04436},
year={2026}
}