Skip to content

David D. Yuan

I’m a robotics researcher at Stanford University and a cofounder of IntBot.

I work on robot learning, studying how robots acquire physical skills and improve them through experience. My research centers on how robots can reason about the physical world and adapt what they learn to new tasks and environments. I’m also interested in how humanoids coordinate movement and manipulation, and how people can teach and work with robots.

Email  /  Scholar  /  GitHub  /  LinkedIn

David D. Yuan holding a pale yellow snake

News

  • Nov 12, 2026OpenWAM received the Best Position Paper award at the CoRL 2026 “Do Robots Need World Models?” workshop.
  • Nov 12, 2026Mulligan received the Best Paper award at the CoRL 2026 “Oops, I Erred” workshop.
  • Oct 06, 2026Our new preprint, OpenWAM, is out, with code and a project page.
  • Oct 05, 2026Mulligan is out: making every round of supervised robot learning count.

Research

An OpenWAM robot lifting bread during a manipulation task

OpenWAM: An Open Framework for Composable World-Action Models

Heng Yu*, David D. Yuan*, Juze Zhang*, Changan Chen, Yao Feng, Michelle Baldonado, Steve Cousins, Li Fei-Fei, Jiajun Wu, Ehsan Adeli

In submission; and

CoRL 2026 Workshop: Do Robots Need World Models?

Best Position Paper

A shared video foundation for studying how prediction, action generation, and reusable dynamics models work together.

project page paper code

BibTeX for OpenWAM: An Open Framework for Composable World-Action Models
@article{yu2026openwam,
  title = {OpenWAM: An Open Framework for Composable World-Action Models},
  author = {Heng Yu and David D. Yuan and Juze Zhang and Changan Chen and Yao Feng and Michelle Baldonado and Steve Cousins and Li Fei-Fei and Jiajun Wu and Ehsan Adeli},
  journal = {arXiv preprint arXiv:2610.07922},
  year = {2026},
  url = {https://arxiv.org/abs/2610.07922}
}

* Equal contribution

Mulligan figure showing diverse initial states, concentrated failures, and targeted data collection

Mulligan: Performance-Guided Data Collection for Efficient On-Robot Learning

Lars Ankile, Perry Dong, Rohan Bhowmik, Aneesh Muppidi, David D. Yuan, Shuran Song, Chelsea Finn

In submission; and

CoRL 2026 Workshop: Oops, I Erred

Best Paper

Directing human-supervised data collection toward failures and unexplored states to improve real-world manipulation more efficiently.

project page paper code

BibTeX for Mulligan: Performance-Guided Data Collection for Efficient On-Robot Learning
@misc{mulligan2026,
  title = {Mulligan: Performance-Guided Data Collection for Efficient On-Robot Learning},
  author = {Lars Ankile and Perry Dong and Rohan Bhowmik and Aneesh Muppidi and David D. Yuan and Shuran Song and Chelsea Finn},
  year = {2026},
  eprint = {2610.05882},
  archivePrefix = {arXiv},
  primaryClass = {cs.RO},
  url = {https://arxiv.org/abs/2610.05882}
}
Robot manipulation frame from the SpeedTuning paper figure

SpeedTuning: Speeding Up Policy Execution with Lightweight Reinforcement Learning

David D. Yuan, Tony Z. Zhao, Kaylee Burns, Chelsea Finn

ICRA, 2025

Learning when to move quickly and when to slow down, accelerating a learned manipulation policy without collecting new demonstrations.

project page paper code

BibTeX for SpeedTuning: Speeding Up Policy Execution with Lightweight Reinforcement Learning
@inproceedings{yuan2025speedtuning,
  title = {SpeedTuning: Speeding Up Policy Execution with Lightweight Reinforcement Learning},
  author = {David D. Yuan and Tony Z. Zhao and Kaylee Burns and Chelsea Finn},
  booktitle = {2025 IEEE International Conference on Robotics and Automation (ICRA)},
  pages = {1184--1192},
  year = {2025},
  doi = {10.1109/ICRA55743.2025.11128753}
}
Visitors interacting with the RoboCrowd robot in a university café

RoboCrowd: Scaling Robot Data Collection through Crowdsourcing

Suvir Mirchandani, David D. Yuan, Kaylee Burns, Md Sazzad Islam, Tony Z. Zhao, Chelsea Finn, Dorsa Sadigh

ICRA, 2025

Conference Best Paper Finalist

Making robot data collection a public, engaging activity through rewards, curiosity, and friendly competition.

project page paper

BibTeX for RoboCrowd: Scaling Robot Data Collection through Crowdsourcing
@inproceedings{mirchandani2025robocrowd,
  title = {RoboCrowd: Scaling Robot Data Collection through Crowdsourcing},
  author = {Suvir Mirchandani and David D. Yuan and Kaylee Burns and Md Sazzad Islam and Tony Z. Zhao and Chelsea Finn and Dorsa Sadigh},
  booktitle = {2025 IEEE International Conference on Robotics and Automation (ICRA)},
  pages = {1392--1399},
  year = {2025},
  url = {https://arxiv.org/abs/2411.01915}
}
Illustration of a Breakout game with a paddle, ball, and colored bricks

A Fast and Accurate Machine Learning Autograder for the Breakout Assignment

Evan Zheran Liu, David Yuan, Ahmed Ahmed, Elyse Cornwall, Juliette Woodrow, Kaylee Burns, Allen Nie, Emma Brunskill, Chris Piech, Chelsea Finn

SIGCSE, 2024

A reinforcement learning agent plays student programs and surfaces bugs for instructors, reducing the work of grading interactive assignments.

project page paper

BibTeX for A Fast and Accurate Machine Learning Autograder for the Breakout Assignment
@inproceedings{liu2024breakout,
  title = {A Fast and Accurate Machine Learning Autograder for the Breakout Assignment},
  author = {Evan Zheran Liu and David Yuan and Ahmed Ahmed and Elyse Cornwall and Juliette Woodrow and Kaylee Burns and Allen Nie and Emma Brunskill and Chris Piech and Chelsea Finn},
  booktitle = {Proceedings of the 55th ACM Technical Symposium on Computer Science Education V. 1},
  year = {2024},
  doi = {10.1145/3626252.3630759}
}

Get in touch

For research, robotics, or a shared curiosity, you can reach me at davidy02 [at] stanford [dot] edu.

Updated November 2026 · Website template by Jon Barron