Selected papers include 'SkillMimic-V2: Learning Robust and Generalizable Interaction Skills from Sparse and Noisy Demonstrations' (SIGGRAPH 2025), 'SkillMimic: Learning Basketball Interaction Skills from Demonstrations' (CVPR 2025), and more high-impact conference publications.
Research Experience
Before postgraduate studies, worked at DH-Robotics as one of the initial team members (the first engineer), leading a team in developing electronics and algorithms for robot grippers and arms. Interned at IDEA Research and Unitree Robotics.
Education
Ph.D. student at HKUST, supervised by Prof. Ping Tan; Master's degree from Peking University; Bachelor's degree from Xidian University.
Background
Research interest: enabling humanoid robots to interact dynamically with objects, environments, and humans. Committed to addressing the data bottleneck in robotics through simulation and reinforcement learning techniques.