PDF攻击让AI文本检测失效，文字内容不变却能骗过所有检测器。
Complete Evasion, Zero Modification: PDF Attacks on AI Text Detection
- 利用PDF图文布局与文本提取顺序的差异，仅调整字符位置。
- 攻击后检测准确率降至50.4%，F1分数归零，但视觉完全无损。
- 揭示了现有检测系统对PDF结构漏洞的脆弱性，适合安全与内容审核研究者。
AI生成文本检测工具在维护内容真实性方面日益重要，但其对抗逃避攻击的鲁棒性仍存疑。本文提出PDFuzz，一种新型攻击方法，利用PDF文档中视觉文本布局与文本提取顺序之间的差异。该方法在保持文本内容完全一致的前提下，通过操控字符位置来打乱提取序列。我们在包含人类与AI生成文本的数据集上，针对ArguGPT检测器进行评估。结果表明，攻击实现完全逃逸：检测准确率从(93.6 ± 1.4)%下降至(50.4 ± 3.2)%，F1分数从0.938 ± 0.014降至0.0，同时视觉保真度保持完美。本工作揭示了当前检测系统在PDF文档结构上的固有漏洞，强调需加强防御机制。代码已公开于https://github.com/ACMCMC/PDFuzz。
原文摘要 · Abstract (English)
AI-generated text detectors have become essential tools for maintaining content authenticity, yet their robustness against evasion attacks remains questionable. We present PDFuzz, a novel attack that exploits the discrepancy between visual text layout and extraction order in PDF documents. Our method preserves exact textual content while manipulating character positioning to scramble extraction sequences. We evaluate this approach against the ArguGPT detector using a dataset of human and AI-generated text. Our results demonstrate complete evasion: detector performance drops from (93.6 $\pm$ 1.4) % accuracy and 0.938 $\pm$ 0.014 F1 score to random-level performance ((50.4 $\pm$ 3.2) % accuracy, 0.0 F1 score) while maintaining perfect visual fidelity. Our work reveals a vulnerability in current detection systems that is inherent to PDF document structures and underscores the need for implementing sturdy safeguards against such attacks. We make our code publicly available at https://github.com/ACMCMC/PDFuzz.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。