Max Robotics — Global Service Robot Directory + US Certification

About

Kuang-Huei Lee is a leading researcher at the intersection of robotics and artificial intelligence, whose work has fundamentally reshaped how robots understand and execute human language. His primary research areas center on grounding large language models (LLMs) in robotic control, enabling machines to translate high-level, natural language instructions into precise physical actions. Lee’s most impactful contribution is the "Do As I Can, Not As I Say" framework (2022, 516 citations), which pioneered a method for grounding LLMs in robotic affordances—effectively teaching robots to consider what actions are physically possible, not just semantically correct. He further advanced the field with the RT-1: Robotics Transformer (2023, 512 citations), a scalable model that leverages diverse, task-agnostic datasets for real-world control, achieving unprecedented generalization across hundreds of tasks. Lee also introduced "Language to Rewards for Robotic Skill Synthesis" (2023, 38 citations), an innovative approach that uses LLMs to define reward functions, allowing robots to learn complex skills through in-context reasoning. His work has been instrumental in bridging the gap between high-level language understanding and low-level motor control, making him a pivotal figure in the development of more capable, intuitive, and safe robotic systems.