HuggingFace Daily Papers（社区热门论文）· HuggingFace Daily Papers（社区热门论文）·· 2026-06-11精选AI 评分76
对抗性重新包装：仅修改呈现层即可欺骗AI同行评审
AI 导读
研究提出对抗性重新包装攻击，在不改动科学证据（方法、实验、数据等）的前提下，仅修改摘要、贡献定位、相关工作、讨论和叙事结构等呈现层内容，并利用AI审稿人反馈进行闭环搜索。在三种主流AI审稿系统上，攻击成功率达75.1%，平均得分提高+1.21/10。策略中，相关工作重定位和分析性讨论扩展等结构性改动效果显著优于表面编辑。分析揭示两种失败模式：AI审稿人更易被亮点打动而非被说服，且会将“看起来解决了限制”与“实际解决”相混淆。研究发布了无污染滚动基准和攻击框架用于测试内容锚定性。
推荐理由
这篇论文戳破一个令人不安的真相：AI 审稿人可以被纯粹的文字包装欺骗，不碰证据就能大幅拉升评分。它把论文呈现本身变成了一枚可优化的攻击面，做学术出版与 AI 评估的人都要正视这个结构性缺陷。
正文
Authors:Xu Yang, Zhizhou Sha, Junbo Li, Jian Yu, Yifan Sun, Matthew Zhao, Jinrui Fang, Xinyue Guo, Yining Wu, Xu Hu, Yifu Luo, Qiang Liu, Zhangyang Wang
Abstract:As AI-generated reviews move from experimental tools into peer-review infrastructure, most robustness concerns have focused on explicit attacks such as hidden instructions and prompt injection. We study a harder and more policy-relevant failure mode: no hidden text, no prompt injection, and no changes to methods, experiments, figures, equations, proofs, or numerical results. The attacker modifies only presentation-level content, such as the abstract, contribution framing, related work, discussion, and narrative structure. We introduce adversarial repackaging: a closed-loop attack that uses AI-reviewer feedback to search for presentation-level revisions while keeping the scientific evidence fixed. Across three mainstream AI reviewers, adversarial repackaging achieves a 75.1% attack success rate and a mean score gain of +1.21/10. The effect is not explained by ordinary prose polishing. We also reveal that strategies that change how the reviewer interprets the paper, such as related-work repositioning and analytical discussion expansion, substantially outperform surface edits such as local polishing, table formatting, and algorithm boxes.
Our analysis reveals two deeper structural failure modes. First, AI reviewers are easier to impress than to convince: highlighting strengths reliably increases perceived merit, while attempts to dissolve weaknesses frequently backfire. Second, AI reviewers can confuse the appearance of addressing a limitation with actually resolving it, allowing unchanged evidence to be reinterpreted as stronger scientific contribution. These results show that the deployment risk is not only malicious hidden instructions, but the emergence of paper presentation itself as an optimization surface. We release a contamination-free rolling benchmark and attack framework for testing whether AI reviewers remain anchored to scientific content under presentation-only edits.
| Comments: | 35 pages, 5 figures |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2606.13044 [cs.CL] |
| (or arXiv:2606.13044v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2606.13044 arXiv-issued DOI via DataCite |
Submission history
From: Xu Yang [view email]
[v1]
Thu, 11 Jun 2026 08:30:18 UTC (1,536 KB)
来源：HuggingFace Daily Papers（社区热门论文） · arxiv.org