I am a VLA Research Team Lead at AIRoA, where I focus on building robust and generalized vision-language-action models. I obtained my Ph.D. degree at UC San Diego in the CSE department under the supervision of Xiaolong Wang. Previously, I was a Research Scientist at NVIDIA Research and a Student Researcher at Google DeepMind, hosted by Kuang-Huei Lee. In the past, I've also worked with Hao Su at UCSD and Masashi Sugiyama at RIKEN-AIP.
I received my B.S. in Electrical Engineering and an M.S. in Computer Science from National Taiwan University. My research interests lie in the fields of reinforcement learning, robotics, and computer vision. Specifically, I am devoted to developing innovative methods for real-world applications. My primary focus is on building robust and generalized vision-language-action models for embodied AI.
Please see my CV for more information. If you would like to know more about my research, please contact me via email, kris.wu [at] airoa.org.
A-0 learns long-horizon, precise bimanual manipulation from 16,151 hours of demonstrations recorded with YUBI, our open-source handheld gripper, with no teleoperation, reinforcement learning or on-robot corrections. It is embodiment-agnostic: the same weights run on a bimanual Franka, TX-G2, OpenArm and UR, and on the Franka it performs ten tasks, such as parts kitting and towel folding, at a site where none of its training data was recorded. Its loss after post-training falls as a power law in hours of handheld data.
[project page] [arXiv] [code] [weights] Technical Report, 2026