Logical Reasoning Challenges
Subbarao emphasizes the importance of logical reasoning in AI, contrasting it with the prevalent focus on deep learning. He highlights that LLMs struggle with deductive closure and transitive closure, revealing fundamental limitations in their reasoning capabilities. Despite their training, these models cannot perform necessary logical inferences, underscoring a critical gap in current AI methodologies.In this clip
From this podcast
Machine Learning Street Talk (MLST)
Prof. Subbarao Kambhampati - LLMs don't reason, they memorize (ICML2024 2/13)
Related Questions