AI RESEARCH
VisualLeakBench: Auditing the Fragility of Large Vision-Language Models against PII Leakage and Social Engineering
arXiv CS.AI
•
ArXi:2603.13385v1 Announce Type: cross As Large Vision-Language Models (LVLMs) are increasingly deployed in agent-integrated workflows and other deployment-relevant settings, their robustness against semantic visual attacks remains under-evaluated -- alignment is typically tested on explicit harmful content rather than privacy-critical multimodal scenarios. We