LLM UNDERSTANDING: 40. Tom GRIFFITHS "Cognitive Science Tools for Understanding Behavior of LLMs"

Описание: COGNITIVE SCIENCE TOOLS FOR UNDERSTANDING THE BEHAVIOR OF LARGE LANGUAGE MODELS

Tom Griffiths

Computer Science, Princeton University

ISC Summer School on Large Language Models: Science and Stakes, June 3-14, 2024

FRI, June 14, 3:30PM-5pm EDT


ABSTRACT: Large language models have been found to have surprising capabilities, even what have been called “sparks of artificial general intelligence.” However, understanding these models involves some significant challenges: their internal structure is extremely complicated, their training data is often opaque, and getting access to the underlying mechanisms is becoming increasingly difficult. As a consequence, researchers often have to resort to studying these systems based on their behavior. This situation is, of course, one that cognitive scientists are very familiar with — human brains are complicated systems trained on opaque data and typically difficult to study mechanistically. In this talk I will summarize some of the tools of cognitive science that are useful for understanding the behavior of large language models. Specifically, I will talk about how thinking about different levels of analysis (and Bayesian inference) can help us understand some behaviors that don’t seem particularly intelligent, how tasks like similarity judgment can be used to probe internal representations, how axiom violations can reveal interesting mechanisms, and how associations can reveal biases in systems that have bee trained to be unbiased.

TOM GRIFFITHS is the Henry R. Luce Professor of Information Technology, Consciousness and Culture in the Departments of Psychology and Computer Science at Princeton University. His research explores connections between human and machine learning, using ideas from statistics and artificial intelligence to understand how people solve the challenging computational problems they encounter in everyday life. Tom completed his PhD in Psychology at Stanford University in 2005, and taught at Brown University and the University of California, Berkeley before moving to Princeton. He has received awards for his research from organizations ranging from the American Psychological Association to the National Academy of Sciences and is a co-author of the book Algorithms to Live By, introducing ideas from computer science and cognitive science to a general audience.

Yao, S., Yu, D., Zhao, J., Shafran, I., Griffiths, T., Cao, Y., & Narasimhan, K. (2024). Tree of thoughts: Deliberate problem solving with large language models. Advances in Neural Information Processing Systems, 36.

Hardy, M., Sucholutsky, I., Thompson, B., & Griffiths, T. (2023). Large language models meet cognitive science: Llms as tools, models, and participants. In Proceedings of the annual meeting of the cognitive science society (Vol. 45, No. 45).