[go: up one dir, main page]
Tianshu Huang

I am a Research Engineer at Meta Superintelligence Labs (MSL), where I work on enabling frontier models with perception and robotics abilities. Before joining Meta, I worked at Assured Robot Intelligence (ARI), which Meta acquired in May 2026. I earned a Master's degree in Computer Science from UC San Diego, where I was a graduate researcher with Xiaolong Wang. Previously, I got a Bachelor's degree in Electrical and Computer Engineering (ECE) from Shanghai Jiao Tong University and a Bachelor's degree in Computer Science (CS) from University of Michigan. I also worked as a research intern at Deep Robotics in 2024 summer.

My research interests include robot learning, humanoid control, and embodied perception.

Google Scholar  /  LinkedIn  /  Twitter  /  Github

Tianshu Huang profile photo

Research
AMO: Adaptive Motion Optimization for Hyper-Dexterous Humanoid Whole-Body Control
Jialong Li*, Xuxin Cheng*, Tianshu Huang*, Shiqi Yang, Ri-Zhao Qiu, Xiaolong Wang
Robotics: Science and Systems (RSS), 2025
project page / arXiv
Humanoid robot performing contact-rich manipulation tasks with HMC
HMC: Learning Heterogeneous Meta-Control for Contact-Rich Loco-Manipulation
Lai Wei*, Xuanbin Peng*, Ri-Zhao Qiu, Tianshu Huang, Xuxin Cheng, Xiaolong Wang
International Conference on Robotics and Automation (ICRA), 2026
project page / arXiv / video
Egocentric human demonstrations and humanoid manipulation examples from In-N-On
In-N-On: Scaling Egocentric Manipulation with In-the-Wild and On-Task Data
Xiongyi Cai*, Ri-Zhao Qiu*, Geng Chen, Lai Wei, Isabella Liu, Tianshu Huang, Xuxin Cheng, Xiaolong Wang
arXiv, 2025
project page / arXiv / code

Other Projects

I trained some quadruped robot locomotion when I was an intern at Deep Robotics.

Quadruped robot parkour Quadruped robot stand manipulation