Jiahang Cao

Hong Kong University of Science and Technology, Shanghai Jiao Tong University

Papers

Total Citations

H-Index

About

Jiahang Cao is an emerging researcher at the forefront of robotics, reinforcement learning, and embodied intelligence, with a focus on advancing the capabilities of legged and dexterous robotic systems. His work bridges cutting-edge machine learning techniques with real-world robotic applications, tackling some of the field's most demanding challenges. In his highly regarded work on fully spiking neural networks for legged robots, Cao explores energy-efficient, biologically inspired architectures as a compelling alternative to conventional deep reinforcement learning methods for quadruped and humanoid locomotion. Complementing this, his research on world model-based perception addresses the critical challenge of data-inefficient visual learning in legged locomotion, enabling robots to navigate complex terrains with greater robustness and adaptability. His development of RoboDexVLM demonstrates a forward-thinking integration of vision-language models with dexterous manipulation planning, pushing beyond the limitations of simplified robotic grasping tasks. Though early in his career — with his most-cited works accumulating citations in 2025 — Cao's research agenda reflects both technical depth and broad ambition, positioning him as a promising contributor to the next generation of intelligent, physically capable robotic systems.