Yichao Liang

Papers

Total Citations

H-Index

About

Yichao Liang is pioneering the frontier of embodied AI, with a focus on enabling robots to learn and adapt generalizable manipulation skills. His work tackles one of the field’s most persistent challenges: creating robotic systems that can robustly handle diverse task configurations, including variations in object shape, density, friction, and external disturbances. In his highly cited 2024 paper, "Rapid Motor Adaptation for Robotic Manipulator Arms," Liang introduces frameworks that allow robots to adapt their motor policies on the fly, a critical step toward deploying robots in unstructured, real-world environments. Complementing this, his work on "VisualPredicator: Learning Abstract World Models with Neuro-Symbolic Predicates for Robot Planning" (2024) proposes a novel neuro-symbolic approach that bridges high-level reasoning with low-level sensorimotor control. By developing a first-order abstraction language, Liang enables robots to form task-specific world models that selectively focus on essential elements, dramatically improving planning efficiency. With his papers quickly accumulating citations and shaping the discourse on robot learning, Liang is establishing himself as a rising leader in creating more intelligent, adaptable, and capable robotic agents.