AI RESEARCH
EDIS: Diagnosing LLM Reasoning via Entropy Dynamics
arXiv CS.LG
•
ArXi:2602.01288v2 Announce Type: replace Entropy-based confidence signals are increasingly leveraged to improve reasoning in large language models (LLMs), yet existing approaches treat confidence as a static quantity -- typically aggregated over tokens. We show that the \emph{temporal evolution} of confidence during generation carries richer information than aggregate statistics alone.