Testing LLMs' Limits
Peter discusses his approach to evaluating large language models, particularly in the medical field. He emphasizes the importance of models being able to disagree and engage in meaningful collaboration, noting that many current models still fall short. With advancements like memory features, there's hope for future improvements, but the journey toward true AGI remains ongoing.In this clip
From this podcast

The Gradient
Peter Lee: Computing Theory and Practice, and GPT-4's Impact
Related Questions