Mechanistic experiments on Gemma 3 27B, Qwen 2.5 7B and Magistral Small 24B show verbal confidence is cached at post-answer positions from answer tokens and captures richer answer-quality information beyond token log-probabilities.
Improving metacog- nition and uncertainty communication in language mod- els.arXiv preprint arXiv:2510.05126, 2025a
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 2verdicts
UNVERDICTED 2roles
background 1polarities
background 1representative citing papers
Meta-d′ and signal detection theory give comparable measures of AI metacognitive sensitivity and risk-sensitive decision regulation, demonstrated on three LLMs.
citing papers explorer
-
How do LLMs Compute Verbal Confidence
Mechanistic experiments on Gemma 3 27B, Qwen 2.5 7B and Magistral Small 24B show verbal confidence is cached at post-answer positions from answer tokens and captures richer answer-quality information beyond token log-probabilities.
-
Measuring the metacognition of AI
Meta-d′ and signal detection theory give comparable measures of AI metacognitive sensitivity and risk-sensitive decision regulation, demonstrated on three LLMs.