Joseph J. Lim

Joseph J. Lim is a faculty member in the Department of Computer Science at Korea Advanced Institute of Science and Technology (KAIST). He leads the Cognitive Learning for Vision and Robotics (CLVR) lab, focusing on research topics such as reinforcement learning, robot learning, world models, visual perception, and symbolic manipulation. His work includes developing intelligent systems for sequential decision-making through perception, action, and reasoning. Lim's research has resulted in multiple paper acceptances at top conferences such as ICML, ICLR, CoRL, and RSS, as well as awards for outstanding contributions in empirical reinforcement learning and robotic learning datasets.

Joseph J. Lim is an Associate Professor at KAIST with expertise in Machine Learning, AI, Reinforcement Learning, and Robotics. His research focuses on developing advanced algorithms for robotic navigation, manipulation, and perception. Lim's work includes deep reinforcement learning for target-driven visual navigation, robotic learning datasets, and multimodal imitation learning. He has contributed to projects such as Open X-Embodiment, Droid, and Furniturebench, which aim to enhance robotic capabilities through large-scale datasets and complex manipulation tasks. His research also explores physics-based learning, object property recognition, and meta-learning techniques for improved generalization in AI systems.