arXiv cs.CY· Grant Shanklin, Ji\v{r}\'i N\v{e}me\v{c}ek, Asia Biega·· 11 小时前AI 评分27
如何应对解释忠实性评估标准之间的分歧
Navigating Disagreement Among Criteria for Evaluating Explanation Faithfulness
AI 导读
针对同一解释是否忠实，现有评估标准可能给出相互矛盾的结论。研究者提出一个忠实性函数框架，将分歧定位在解释对模型作出何种断言、哪些比较与断言相关、以及如何聚合这些比较三个环节，并据此重构现有评估标准。以不利行动通知为案例的研究显示，具体场景可论证部分选择、缩小标准间的取舍空间。
正文 · 原文
Abstract:Faithfulness requires that algorithmic explanations accurately reflect the model's behavior. Yet existing operationalizations of faithfulness can disagree about whether the same explanation is faithful, leaving developers and auditors to choose among conflicting evaluations. In this paper, we investigate disagreement among evaluation criteria and offer a practical path forward for evaluating explanation faithfulness. We first identify how operationalizations of faithfulness differ. We introduce a faithfulness functional that locates their disagreements in what an explanation claims about the model, which comparisons bear on that claim, and how agreement across those comparisons should be aggregated, and we reconstruct existing evaluation criteria within it. We argue these reconstructions support understanding faithfulness as a common though incompletely specified relation between the model and explanation, so the differences reflect choices left open by the general requirement. Justifying an operationalization then comes down to justifying these choices, which, on our view, draws on explanatory, institutional, and normative considerations. We support this argument with a case study of adverse action notices, reconstructing the evaluation implicit in an industry-proposed procedure and identifying which of its choices the setting justifies and which remain open. Our work gives auditors and practitioners a way to navigate disagreeing evaluations of faithfulness. It also explains why the literature should expect multiple operationalizations, while showing how concrete settings can narrow the choice among them. More broadly, our analysis shows how a shared requirement of faithfulness can support contextually justified evaluations despite disagreement among existing metrics.
| Subjects: | Machine Learning (cs.LG); Computers and Society (cs.CY) |
| Cite as: | arXiv:2611.06846 [cs.LG] |
| (or arXiv:2611.06846v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2611.06846 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Grant Shanklin [view email]
[v1]
Wed, 4 Nov 2026 18:03:15 UTC (164 KB)
来源：arXiv cs.CY · arxiv.org