4 papers
The Computational Basis of Confidence in Large Language Models
Dharshan Kumaran, Viorica Patraucean, Maks Ovsjanikov +2
Reliable confidence -- the probability that a model's own answer is correct -- is essential for the trustworthy deployment of language models. Existing work has largely evaluated c…
Closing the Loop to Discover Psychological Theories with an Automated Cognitive Scientist
Akshay K. Jagadish, Younes Strittmatter, Nori Jacoby +5
Across the sciences, autonomous systems are increasingly being used in closed-loop discovery, proposing new theories and designing and running experiments to test them. This approa…
ATLAS: Active Theory Learning for Automated Science
Noémi Éltető, Nathaniel D. Daw, Kimberly L. Stachenfeld +1
Advancing scientific understanding through mechanistic modeling requires posing the right experimental questions to yield maximally informative data. To automate this pursuit withi…
How LLMs Detect and Correct Their Own Errors: The Role of Internal Confidence Signals
Dharshan Kumaran, Viorica Patraucean, Simon Osindero +2
Large language models can detect their own errors and sometimes correct them without external feedback, but the underlying mechanisms remain unknown. We investigate this through th…