Self-assisting and Cooperative Large Language Models
Abstract:
Large Language Models (LLMs) have emerged as powerful tools in AI, yet their effectiveness in addressing complex problems often encounter limitations. This talk aims to shed light on two pivotal questions: (1) can LLMs solve intricate algorithmic problems (e.g., problems in Olympiad Informatics)? (2) how can LLMs effectively learn from and rectify their errors in complex reasoning tasks? We will introduce the ALGO method which uses self-generated oracle as a guide for algorithmic program generation. Complementing this, we introduce SALAM method to learn cooperative study assistants for LLMs to correct mistakes. These advancements highlight significant progress in enhancing LLMs' problem-solving abilities.
Bio:
Lei Li is an Assistant Professor in Language Technology Institute at Carnegie Mellon University. He received Ph.D. from Carnegie Mellon University School of Computer Science in 2011. He is a recipient of ACL 2021 Best Paper Award, CCF Young Elite Award in 2019, CCF distinguished speaker in 2017, Wu Wen-tsün AI prize in 2017, and 2012 ACM SIGKDD dissertation award (runner-up), and is recognized as Notable Area Chair of ICLR 2023. Previously, he was a faculty member at UC Santa Barbara. Prior to that, he was the Founding Director of ByteDance AI Lab and led its research in NLP, ML, Robotics, and Drug Discovery. He launched ByteDance’s machine translation system VolcTrans and AI writing system Xiaomingbot, and many of his algorithms have been deployed in products (Toutiao, Douyin, Tiktok, Lark), serving over a billion users.