Auditing Question-Order Effects in Large Language Models with the QQ Equality: Mechanism Characterization and a Saturation Caveat
用QQ等式审计大语言模型中的问题顺序效应:机制表征与饱和警告
机构 * Dankook University(韩国檀国大学)
专题命中 评测与基准 :large language model(title,abstract);language model(title,abstract);LLM(abstract);分类 cs.CL、cs.AI
AI总结 研究大语言模型中问题顺序效应,将QQ等式发展为审计标准,理论上表征满足该等式的机制类别,方法上开发审计管道,实验发现强制二元下一个token对数概率不适用于分布级QQ审计,建议进行饱和诊断。
Comments v2: major revision. Five restructured findings separating order sensitivity, QQ imbalance, and residual contextuality; two-layer discriminant table; pipeline figure and per-item joint table; envelope pooling and certified Gamma bounds specified; retrospective batch-1 G3 re-validation (all verdicts preserved). No new pilot measurements