A Reference-Free Score for Detecting Silent Reasoning Failures in Large Language Models

본문 미리보기

arXiv:2607.26102v1 Announce Type: new Abstract: Mathematical chain of thought (CoT) evaluation is commonly reduced to whether the final answer matches a reference. This conflates producing a correct conclusion with producing a valid derivation an invalid chain can accidentally reach the right answer, while a valid calculation can be followed by a transcription error. We call this mismatch the reasoning answer consistency gap. This framework paper introduces the Reasoning Answer Faithfulness Sco