arXiv cs.AI· Yuanchuan Guo, Jiashun Jin, Zheng Tracy Ke, Bingcheng Sui·· 12 小时前AI 评分38
AI4Math 淘金热 II：AI 辅助证明有哪些特征？
The Gold Rush in AI4Math, II: What are signatures of AI-assisted Proofs?
AI 导读
研究构建了包含 8,281 篇论文的五个队列，并提出 ProofTrace 框架，将论文表示为即时证明支撑图（GIPS）以提取可解释的数学组织特征。研究发现 AI 生成证明存在反复出现的特征，如更少引用外部结果、更依赖中间结论、图连通性改变，且这些特征在研究领域、稿件长度或图规模控制下依然存在。
正文 · 原文
Abstract:The growing use of AI in mathematics raises a question: can we detect AI involvement in proofs, separately from AI assistance with writing? Historical arXiv papers, recent papers disclosing substantive AI use, and manuscripts released by OpenAI provide a rich empirical basis for studying this question. However, several challenges remain: lack of matched case-control pairs; potential incompleteness of disclosure; and lack of ready-to-use features designed to capture signals of AI involvement in proofs.
To address these challenges, we construct five cohorts comprising 8,281 papers by combining existing manuscripts with those we produced using generative AI. These cohorts span different levels of AI involvement and include matched pairs of papers addressing the same research problems. We then introduce ProofTrace, a framework that represents each paper as a graph of immediate proof support (GIPS) and extracts interpretable features of mathematical organization.
Our study yields three discoveries. First, principal-component projections of our features reveal a qualitative gradient of AI involvement, where AI-reworded papers remain close to human papers, the two AI-generated cohorts depart in broadly shared directions, and papers disclosing substantive AI use occupy intermediate positions which shifts toward AI-generated cohorts over time. Second, we identify recurring signatures of AI-generated proofs, such as fewer externally cited results, greater reliance on intermediate conclusions, and altered graph connectivity; many of these signatures persist under controls for research field, manuscript length, or graph size. Finally, in a separate detection experiment, ProofTrace distinguishes human-written papers from those disclosing substantive AI use. Its performance remains stable under AI rewording, whereas the performance of state-of-the-art text-based detectors declines sharply.
| Comments: | 31 pages, 15 figures |
| Subjects: | Artificial Intelligence (cs.AI); Applications (stat.AP) |
| Cite as: | arXiv:2611.08412 [cs.AI] |
| (or arXiv:2611.08412v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.08412 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Bingcheng Sui [view email]
[v1]
Thu, 5 Nov 2026 18:17:18 UTC (4,714 KB)
来源：arXiv cs.AI · arxiv.org