Yonghyeon Lee is a Postdoctoral Associate at the MIT Biomimetic Robotics Lab, working with Prof. Sangbae Kim. Previously, he was an AI Research Fellow at the Korea Institute for Advanced Study (KIAS) from March 2023 to February 2025. He earned his Ph.D. from SNU Robotics Lab under the guidance of Prof. Frank C. Park (March 2018 – February 2023). Before that, he completed his B.S. in Mechanical Engineering and Physics at SNU.
Talk Title: Toward General Physical Intelligence: Why Physics, Control, and Geometry Still Matter in the Era of Data
Abstract:
Robotics is receiving unprecedented attention, driven by the success of large-scale data-driven models such as GPT and the growing belief that similar scaling principles may lead to general-purpose robot intelligence. In robotic manipulation, this has fueled major investments in vision-language-action models, world models, and large-scale generative imitation learning. While some may argue that sufficient data and computation can replace physics modeling, control theory, and geometric reasoning, this talk revisits why these foundations remain essential for achieving general physical intelligence. Using robotic grasping as a case study, I will discuss how learning, physics, control, and geometry can be integrated to build robotic systems that are more robust, adaptive, and generalizable in the physical world.
RPL Summer School 2026 