Limitations of LLMs
François highlights the inherent limitations of large language models, emphasizing that many issues, such as their autoregressive nature, have persisted since their inception. Despite attempts to patch certain problems through reinforcement learning from human feedback, new variations of these issues continue to emerge. The conversation underscores a lack of significant progress in addressing these fundamental challenges.In this clip
From this podcast
Machine Learning Street Talk (MLST)
It's Not About Scale, It's About Abstraction - Francois Chollet
Related Questions