معرفی
Yunzhu Li is an Assistant Professor of Computer Science at Columbia University, leading the Robotic Perception, Interaction, and Learning Lab (RoboPIL). Previously, she held positions at UIUC and Stanford, and completed her PhD at MIT under advisors Antonio Torralba and Russ Tedrake, with a bachelor's from Peking University. Her research focuses on robotic manipulation, embodied intelligence, and multi-modal perception, emphasizing physics-inspired models and foundation models for generalizable robot learning. RoboPIL develops structured world models for deformable objects, integrates multi-modal sensing (vision, touch, audio), and explores long-horizon embodied interactions.
Key research directions include structured world models for manipulable objects, embodied intelligence through foundation models, and multi-modal perception fusion. Recent talks highlight advancements in robot learning, including workshops at ICRA, NUS, and RSS. Her lab actively engages students in PhD, Master's, and undergraduate programs, emphasizing technical strength and self-motivation.
Scientific awards include the Best Paper Award at ICRA 2024 Workshop and Sony Faculty Innovation Award. Her work has been featured in venues like Science, Nature Electronics, and IEEE journals. Professional service includes organizing workshops on structured world models and embodied agents, and serving on program committees for CVPR, ICLR, and RSS.
