Jayjun Lee
Papers
2
Total Citations
6
H-Index
2
About
Jayjun Lee is a rising force in robotics and embodied AI, whose work bridges the critical gap between perception, manipulation, and autonomous recovery. His research centers on two pivotal challenges: enabling robots to understand their physical interactions with the world through multimodal sensing, and equipping them with the ability to learn from and correct their own mistakes. In his highly cited work on **ViTaSCOPE**, Lee introduced a novel visuo-tactile implicit neural representation that allows robots to simultaneously estimate an object’s in-hand pose and external contact points. This breakthrough, which fuses high-resolution tactile data with vision, is foundational for precise dexterous manipulation. Complementing this, his **RACER** framework tackles the fragility of imitation learning by proposing a scalable pipeline for generating language-guided failure recovery policies. This work empowers robots to not only execute tasks but to autonomously diagnose and correct errors using rich semantic feedback. With both papers published in 2025 already accumulating citations, Lee’s contributions are rapidly shaping the future of robust, perceptive, and self-correcting robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2