Jason Jingzhou Liu
Jason Jingzhou Liu is a PhD student in robotics at Carnegie Mellon University's School of Computer Science, where he is advised by Deepak Pathak and Ruslan Salakhutdinov. His research focuses on robot learning for manipulation, and he is best known as the lead author of the FACTR line of work on force-aware policy learning for contact-rich manipulation, including FACTR (RSS 2025) and FACTR 2. Prior to his doctoral studies, he contributed to dexterous manipulation and simulation projects at NVIDIA and the University of Toronto.
“NEXT (Neural External Torque Estimation) trains a small LSTM on 10 minutes of free-motion data — no contact, no labels, no force sensor required — and achieves external torque estimation accuracy of 0.547 ± 0.348 Nm L1 error during contact on a Franka arm”
Source→“Correspondence to: Jason Jingzhou Liu, liujason@cmu.edu”
Source→AI-extracted from podcast / newsletter / paper summaries. May contain errors.