Reasoning in AI
Subbarao discusses the evolving landscape of large language models (LLMs) and the hidden costs associated with reasoning tokens. He highlights the tension between general-purpose systems and specialized solvers, emphasizing the importance of reasoning for accuracy despite its computational complexity. The conversation also touches on the emerging trend of compound AI systems, which aim to enhance planning and scheduling capabilities.In this clip
From this podcast
Machine Learning Street Talk (MLST)
Subbarao Kambhampati - Do o1 models search?
Related Questions