LLMs start overconfident, then swing underconfident when criticized, Nature MI paper finds
ValerioCapraro · x · 2026-09-23
A Nature Machine Intelligence paper identifies two competing biases in LLMs: merely seeing their own earlier answer inflates confidence (a consistency-preserving bias), while criticism flips them from overconfident to underconfident — with direct implications for self-evaluation and self-consistency pipelines.
Related event: Nature MI Study Finds LLMs Overconfident, Then Overcorrect(2 posts)→
More from Models
- GPT-6 Sol and Luna already usable in Codex, early user reports — airesearch12 · 2026-09-23
- GPT-6 Sol scores slightly below GPT-5.6 Sol on DeepSWE, only cheaper — Angaisb_ · 2026-09-23
- Matt Shumer on Opus 5.5: 'feels like a much smarter Opus 4.6' — mattshumer_ · 2026-09-23
- GPT-6 Sol claimed to cost 50% less than GPT-5.6 Sol — cedric_chee · 2026-09-23
- Four frontier models in days: Grok 4.7, Opus 5.5, GPT-6 Sol and Luna — msg · 2026-09-23
- OpenAI reportedly rolling out GPT-6 Sol and GPT-6 Luna on ChatGPT, Codex, and APIs — testingcatalog · 2026-09-23