How LLMs Detect and Correct Their Own Errors: The Role of Internal Confidence Signals

发布

2026年04月27日

采集 2026年04月27日 06:32

学术前沿 7.5 分 — Insightful mechanistic study of error detection via second-order confidence models. Bridges decision neuroscience and LLM behavior.

原文： arxiv.org

评分 7.5 · 来源： · 发布于 2026-04-27

评分依据：Insightful mechanistic study of error detection via second-order confidence models. Bridges decision neuroscience and LLM behavior.

PermaFrost-Attack: Stealth Pretraining Seeding(SPS) for planting Logic Landmines During LLM Training

Estimating Tail Risks in Language Model Output Distributions