发表机构
The University of Texas Rio Grande Valley; Florida International University（德克萨斯大学里奥格兰德河谷分校; 佛罗里达国际大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
CleanScore通过阴性对照与敏感性边界进行黑盒基准审计，区分能力与数据暴露，并量化表面形式膨胀的界限。
AI 中文摘要
公开基准分数可能反映模型能力、对问题的先前接触，或两者兼有，且对于大多数模型而言，训练数据是未知的。我们提出CleanScore，一种仅使用评分输出的黑盒审计方法。每个基准问题作为一个父条目，包含一个公开形式及两个独立编写的新鲜形式，保留其数字、事实和答案。该审计报告公开形式优势的区间而非结论，并提供一个具有明确传输半径的私有阴性对照库，以区分暴露与普通形式不匹配。一项注册的受控暴露实验能检测到植入的暴露，并在新鲜形式暴露下保持沉默。一项对五个开放模型在200个GSM8K和200个ARC-Challenge条目上的注册审计未发现与暴露一致的优势，将表面形式膨胀限制在五分以下。注册的阳性对照随后限定了此类零结果可能意味着什么。泄露一个条目会使模型从未见过的改写版本的准确率提高，几乎与泄露措辞本身一样多，使得52%至110%的效果对改写审计不可见。在ARC上，一个植入的49分优势显示出-0.020的可观察差距，且约20分在改写题干和选项后仍然存在，跨越四个训练种子。表面形式零结果所限定的范围远小于短语污染审计所暗示的。
英文摘要
Public benchmark scores may reflect skill, prior exposure to the questions, or both, and for most models the training data are unknown. We present CleanScore, a black-box audit using scored outputs only. Each benchmark question becomes a parent item with one public form and two independently written fresh forms preserving its numbers, facts and answer. The audit reports an interval for the public-form advantage rather than a verdict, and a private negative-control bank with an explicit transport radius separates exposure from ordinary form mismatch. A registered controlled-exposure experiment detects planted exposure and stays quiet under fresh-form exposure. A registered audit of five open models on 200 GSM8K and 200 ARC-Challenge items finds no exposure-consistent advantage, bounding surface-form inflation below five points. Registered positive controls then bound what such a null can mean. Leaking an item raises accuracy on paraphrases the model never saw almost as much as on the leaked wording, leaving 52% to 110% of the effect invisible to a paraphrase audit. On ARC a planted 49-point advantage shows an observable gap of -0.020, and about 20 points survive rewriting stem and options, across four training seeds. A surface-form null bounds far less than the phrase contamination audit implies.
Comments43 pages, 15 figures, 21 tables. Item bank, code and pre-registrations released
