21 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…
SeeSE3: Emergence of 3D Space in Vision Features
Caroline Chen, Sayna Ebrahimi, Fedor Kitashov +4
In this paper, we ask whether vision foundation models construct representations that reflect the intrinsic properties of 3D Euclidean space. Unlike previous works that probe 3D aw…
Gen4U: Unifying Video Generation and Understanding via Diffusion
Michael King, Aravindh Mahendran, Matthew Koichi Grimes +5
Prior work suggests that diffusion representations capture low-level geometry but struggle with high-level semantics. We demonstrate that state-of-the-art video diffusion models ov…
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…
Causal Evidence that Language Models use Confidence to Drive Behavior
Dharshan Kumaran, Nathaniel Daw, Simon Osindero +2
Metacognition -- assessing the quality of one's own cognitive performance -- guides adaptive behavior across species. Substantial research demonstrates that confidence signals can…
How do LLMs Compute Verbal Confidence
Dharshan Kumaran, Arthur Conmy, Federico Barbero +3
Verbal confidence -- prompting LLMs to state their confidence as a number or category -- is widely used to extract uncertainty estimates from black-box models. However, how LLMs in…