多模态（推理）LLM 能否检测文档篡改？

View PDF HTML (experimental)

Abstract:Document fraud poses a significant threat to industries reliant on secure and verifiable documentation, necessitating robust detection mechanisms. This study investigates the efficacy of state-of-the-art multi-modal large language models (LLMs)-including OpenAI O1, OpenAI 4o, Gemini Flash (thinking), Deepseek Janus, Grok, Llama 3.2 and 4, Qwen 2 and 2.5 VL, Mistral Pixtral, and Claude 3.5 and 3.7 Sonnet-in detecting fraudulent documents. We benchmark these models against each other and prior work on document fraud detection techniques using a standard dataset with real transactional documents. Through prompt optimization and detailed analysis of the models' reasoning processes, we evaluate their ability to identify subtle indicators of fraud, such as tampered text, misaligned formatting, and inconsistent transactional sums. Our results reveal that top-performing multi-modal LLMs demonstrate superior zero-shot generalization, outperforming conventional methods on out-of-distribution datasets, while several vision LLMs exhibit inconsistent or subpar performance. Notably, model size and advanced reasoning capabilities show limited correlation with detection accuracy, suggesting task-specific fine-tuning is critical. This study underscores the potential of multi-modal LLMs in enhancing document fraud detection systems and provides a foundation for future research into interpretable and scalable fraud mitigation strategies.

Comments:	arXiv admin note: text overlap with arXiv:2503.20084
Subjects:	Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL)
Cite as:	arXiv:2508.11021 [cs.CV]
	(or arXiv:2508.11021v2 [cs.CV] for this version)
	https://doi.org/10.48550/arXiv.2508.11021 arXiv-issued DOI via DataCite

Submission history

From: Simiao Ren [view email]
[v1] Thu, 14 Aug 2025 18:57:07 UTC (328 KB)
[v2] Sat, 24 Oct 2026 21:36:31 UTC (328 KB)