2 papers
cs.CL2026
Safe Alone, Unsafe Together: Safeguarding Against Implicit Toxicity When Benign Images Combine
Jiaxian Lv, Shiyao Cui, Yingkang Wang +3
Multi-image content has become an increasingly prevalent form of visual communication in social media, giving rise to a new safety issue, multi-image implicit toxicity (MIIT), wher…
cs.CR2026
Observable Channels, Not Just Storage: Evaluating Privacy Leakage in LLM Agent Pipelines
Tao Huang, Chen Hou, Guosen Wu +1
Privacy leakage in LLM agents is often studied through individual storage or execution components, such as memory modules, retrieval pipelines, or tool-mediated artifacts. However,…