Yuke Zhu

Papers

Total Citations

H-Index

About

Yuke Zhu is an influential researcher at the forefront of robot learning and embodied artificial intelligence. His work focuses on developing systems that enable robots to acquire, retain, and transfer knowledge across diverse tasks — a critical challenge in building truly adaptable autonomous agents. His notable contribution, **LIBERO** (2023), introduces a rigorous benchmarking framework for evaluating lifelong robot learning, specifically targeting how robots can transfer knowledge across sequential tasks without catastrophic forgetting. This work addresses one of the most pressing open problems in robotics: enabling machines to accumulate experience much like humans do, rather than learning each task from scratch. Already accumulating 15 citations shortly after publication, LIBERO has quickly become a reference point for researchers developing continual and transfer learning methods in robotics. Zhu's research sits at the intersection of machine learning, computer vision, and robotics, pushing boundaries in simulation-to-real transfer and imitation learning. His contributions provide the community with standardized tools to measure progress objectively, accelerating the development of next-generation robots capable of lifelong, generalizable learning in complex real-world environments.