Adam Kan

Papers

Total Citations

273

H-Index

About

Adam Kan is a leading researcher at the intersection of robotics and artificial intelligence, with a primary focus on enabling robots to provide personalized physical assistance in everyday environments. His most impactful contribution is the development of **TidyBot**, a system that leverages large language models (LLMs) to allow household robots to learn and generalize user preferences for tidying up. The core innovation addresses a fundamental challenge: how can a robot, after observing a user put away a few items, infer a consistent rule (e.g., "t-shirts go in the drawer, not on the shelf") and apply it to all future objects? Kan’s work demonstrates that LLMs can effectively reason about these contextual preferences, enabling robots to personalize cleanup without explicit programming for every scenario. The seminal 2023 paper on TidyBot has rapidly accumulated over 270 citations, underscoring its significance in the field. This achievement has positioned Kan as a key figure in the push toward more intuitive and adaptable home robots, bridging the gap between general-purpose AI and the nuanced, individual needs of human users.