Jinhan Li

Papers

Total Citations

H-Index

About

Jinhan Li is a rising roboticist whose work pushes the boundaries of legged locomotion and manipulation. His research centers on enabling robots—particularly quadrupeds and humanoids—to perform complex, agile, and long-horizon tasks that were previously the domain of specialized, expensive platforms. Li’s most cited work, "Learning Agile Bipedal Motions on a Quadrupedal Robot" (14 citations), demonstrates a groundbreaking approach: using a lightweight, cost-effective quadruped to achieve human-like bipedal agility, unlocking new possibilities for versatile robot design. He further extends this capability in "Long-horizon Locomotion and Manipulation on a Quadrupedal Robot with Large Language Models" (2 citations), integrating LLMs to imbue robots with high-level semantic reasoning for multi-step tasks. In "OKAMI: Teaching Humanoid Robots Manipulation Skills through Single Video Imitation" (2 citations), Li tackles skill acquisition from minimal data, introducing object-aware retargeting to enable humanoids to learn manipulation from a single RGB-D video. This work highlights his focus on data efficiency and generalization. With a citation count already growing, Li is recognized for bridging the gap between low-level control and high-level planning, making advanced robotic behaviors more accessible and practical.