Beyond Visual Evidence: Revealing and Mitigating Relational Privacy Leakage in Document MLLMs

Computer Science > Computer Vision and Pattern Recognition [Submitted on 13 Aug 2026] Beyond Visual Evidence: Revealing and Mitigating Relational Privacy Leakage in Document MLLMs Beining Xu, Hairui Wang, Jiaxin Wang, Changsheng Chen, Anirban Chakraborty While the privacy risks of multimodal large language models (MLLMs) have drawn significant attention, the unique vulnerabilities of domain-specific MLLMs remain largely underexplored. Focusing on document understanding MLLMs for identity document processing, this paper investigates the privacy issues inherent in Key Information Extraction (KIE) tasks. We reveal that when input images lack sufficient visual evidence, these models often rely on memorized field relations from training data to infer missing content, thereby leaking multiple correlated fields containing sensitive personal information. To mitigate this risk, we make three key this http URL, we propose the Dynamic Relational Unlearning Framework (DRUF) which comprises a Relational Decoupling Unlearning (RDU) module and a dynamic set update mechanism. It suppresses the leakage of high-risk field pairs while preserving KIE this http URL, we introduce DocPrivacyBench, a novel benchmark to systematically evaluate a model's susceptibility to privacy leakage under conditions of absent or minimal visual this http URL, we evaluate three MLLMs and six unlearning methods using this benchmark, assessing both post-unlearning leakage suppression and utility this http URL results demonstrate that existing MLLMs consistently exhibit privacy leakage when visual evidence is scarce, particularly on noisier datasets. In contrast, DRUF outperforms the strongest baseline by improving leakage suppression by 4.8 percentage points, effectively mitigating privacy risks while maintaining robust document information extraction performance. Comments: ACM mm 2026 Subjects: Computer Vision and Pattern Recognition (cs.CV); Cryptography and Security (cs.CR); Multimedia (cs.MM) Cite as: arXiv:2608.12911 [cs.CV]   (or arXiv:2608.12911v1 [cs.CV] for this version)   https://doi.org/10.48550/arXiv.2608.12911 Focus to learn more Related DOI: https://doi.org/10.1145/3767308.3835901 Focus to learn more Submission history From: Beining Xu [view email] [v1] Thu, 13 Aug 2026 07:53:53 UTC (2,903 KB) Access Paper: HTML (experimental) view license Current browse context: cs.CV < prev   |   next > new | recent | 2026-08 Change to browse by: cs cs.CR cs.MM References & Citations NASA ADS Google Scholar Semantic Scholar Export BibTeX Citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Demos Related Papers About arXivLabs Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)