Lucy Xiaoyang Shi
Papers
4
Total Citations
88
H-Index
3
About
Lucy Xiaoyang Shi is a leading researcher at the intersection of robotics, imitation learning, and vision-language-action (VLA) models, with a focus on enabling generalist robots to operate autonomously in complex, real-world environments. Her work tackles the critical challenge of moving robotic manipulation beyond controlled lab settings into unstructured domains like surgery and open-world tasks. Shi’s major contributions include pioneering hierarchical frameworks for autonomous surgery, such as SRT-H, which uses language-conditioned imitation learning to achieve dexterous, long-horizon manipulation on variable human tissue (70 citations). She has also advanced foundational methods in imitation learning, notably through waypoint-based approaches that mitigate compounding errors in behavioral cloning, and co-developed the π₀ and π₀.₅ VLA flow models, which demonstrate remarkable open-world generalization for general robot control. Her research has rapidly gained traction, with over 88 total citations across her most-cited papers, reflecting its immediate impact on the field. By bridging language, vision, and action, Shi is shaping the future of autonomous robotics, making her work essential reading for students and researchers pursuing general-purpose robot intelligence.
Research Focus
Key Achievements
Top Papers
- 1
- 2Waypoint-Based Imitation Learning for Robotic Manipulation8 citations · 2023
- 3$π_0$: A Vision-Language-Action Flow Model for General Robot Control8 citations · 2024
- 4$π_{0.5}$: a Vision-Language-Action Model with Open-World Generalization2 citations · 2025