World Models Explored
Jacob discusses the complexities of world models in neural sequence models, highlighting the diverse interpretations of the term. He draws intriguing parallels between human and scientific models and the representations within language models. The conversation also touches on the findings from the Othello GPT study, emphasizing the nuanced understanding of heuristics and decision rules in AI systems.In this clip
From this podcast
The Gradient
Jacob Andreas: Language, Grounding, and World Models
Related Questions