Selective Disclosure: Controlling Information Leakage in DocVQA Explanations

Kangsoo Jung, Mohamed Ali Souibgui, Changkyu Choi, Catuscia Palamidessi

Keywords: XAI, Document intelligence, Information leakage

TL;DR: Visual explanations in DocVQA can leak sensitive information; we prevent this with a policy-aware, explanation-level masking framework that preserves interpretability.

Abstract: Explainable document visual question answering systems improve transparency by visualizing document regions relevant to a query, but they can unintentionally expose sensitive information through explanation outputs, even when textual answers are restricted. We analyze privacy risks arising from coarse explanations and adversarial jailbreaking queries. To mitigate this, we propose a policy-aware visual explanation sanitization framework based on a Role-centric Attribute-Based Access Control (RABAC) model, combining document structure analysis and line-level localization to mask restricted regions. Our method is model-agnostic, supports dynamic policies, and significantly reduces sensitive information leakage while preserving interpretability.

Submission Number: 131