Detecting Unfaithful Chain-of-Thought via Circuit-Guided Internal-External Discrepancy 文章

ArXiv CS.AI2026-05-26NEWSen作者: Xu Shen, Zhen Tan, Song Wang, Pingjun Hong, Rui Miao, Xin Wang, Tianlong Chen
大语言模型
查看原文 →
关系图谱

Detecting Unfaithful Chain-of-Thought via Circuit-Guided Internal-External Discrepancy · 相关技术

相关技术

ORMODEPLALLMlanguage modelchains of thoughtUCTStraight-Through EstimatorStanReferring expression comprehension (REC)Parts-of-Speech (POS) tagsOWSLMMlarge language modelsForFFIComposed image retrievalchain-of-thoughtCoTCISARGANN