Published Feb 22, 2024
Cameron Jones & Sean Trott: Understanding, Grounding, and Reference in LLMs
Cameron Jones and Sean Trott delve into the intricacies of language development in large language models (LLMs), examining language ambiguity, grounding, and referential understanding while questioning the construct of Theory of Mind. They explore human benchmarks through models' performances and interpret complex cognitive mechanics, shedding light on the true interpretative and cognitive capabilities of LLMs.
Topics covered
Popular Clips
Episode Highlights
Related Episodes
Jacob Andreas: Language, Grounding, and World Models
Answers 383 questions
Subbarao Kambhampati: Planning, Reasoning, and Interpretability in the Age of LLMs
Answers 383 questions
Harvey Lederman: Propositional Attitudes and Reference in Language Models
Answers 383 questions
Raphaël Millière: The Vector Grounding Problem and Self-Consciousness
Answers 383 questions
Linus Lee: At the Boundary of Machine and Mind
Answers 383 questions
Kyunghyun Cho: Neural Machine Translation, Language, and Doing Good Science
Answers 383 questions
Ed Grefenstette: Language, Semantics, Cohere
Answers 383 questions
Ted Underwood: Machine Learning and the Literary Imagination
Answers 383 questions
Tal Linzen: Psycholinguistics and Language Modeling
Answers 383 questions
Ted Gibson: The Structure and Purpose of Language
Answers 383 questions
Stevan Harnad: AI's Symbol Grounding Problem
Answers 383 questions
Catherine Olsson and Nelson Elhage: Anthropic, Understanding Transformers
Answers 383 questions
Zachary Lipton: Where Machine Learning Falls Short
Answers 383 questions
Christopher Manning: Linguistics and the Development of NLP
Answers 383 questions
Terry Winograd: AI, HCI, Language, and Cognition
Answers 383 questions