Language Models' Interpretability
Subbarao discusses how Language Models (LLMs) can provide various linguistic representations, showcasing their ability to generate outputs in different forms. He highlights the challenge of distinguishing between human and LLM-generated content based on syntactic characteristics rather than content, emphasizing the shift towards evaluating form over substance in text generation.In this clip
From this podcast
The Gradient
Subbarao Kambhampati: Planning, Reasoning, and Interpretability in the Age of LLMs
Related Questions