arXiv cs.CL· Yixuan Huang, Danish Pruthi, Bhuwan Dhingra·· 2 天前AI 评分31
面向数学密集型科学文本的多模态 AI 检测
Multimodal AI Detection for Math-Heavy Scientific Text
AI 导读
研究者从 2022 年前的 arXiv LaTeX 源码中选取人类撰写的证明与散文段落，用 15 个现代 LLM 在同一论文语境下重新生成，构建了 LaTeX 源码、PDF 渲染图像和 PDF 提取文本三种表示形式的配对数据。
正文 · 原文
Abstract:As large language models become increasingly capable of generating, and assisting with, scientific writing, there is a need for greater transparency and disclosure around the use of AI in research. Consequently, a veritable subfield of research and a number of commercial services aim to flag AI-generated content automatically. However, existing detectors which rely only on text extracted from PDFs fail to benefit from the rich visual information contained in the formatting of math-heavy scientific content. To study this, we select human-written proof and prose passages from pre-2022 arXiv LaTeX sources and re-generate them using 15 modern LLMs within the context of the same paper. This leaves us with coupled human and AI outputs in three different representations: the LaTeX source, a rendered image of the PDF, and the text extracted from the PDF. We find that the accuracy of the state-of-the-art Pangram 4 drops significantly as the mathematical content in a passage increases, especially when evaluated on text as opposed to LaTeX. However, this is not an inherent limitation of detecting AI-generated math: finetuning Qwen-3.5-27B based detectors on this domain recovers the gap, with LaTeX inputs performing significantly better than text since they preserve more formatting information. Since LaTeX sources are rarely available in practice, we also train SnapJudge: an image-based detector operating on screenshots of both prose and proofs from papers. At 0.1% false positive rate, SnapJudge achieves a detection rate of 92% on proofs and 93% on prose generated by frontier models, substantially outperforming both Pangram 4 and our text-based detector. We release our data, models and code at this https URL.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.21062 [cs.CL] |
| (or arXiv:2610.21062v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.21062 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yixuan Huang [view email]
[v1]
Thu, 15 Oct 2026 01:10:14 UTC (489 KB)
来源：arXiv cs.CL · arxiv.org