Meng Xu

GONDAR is a novel sink-centric fuzzing framework that systematically leverages vulnerability-specific knowledge to discover Java security flaws, significantly outperforming state-of-the-art fuzzers.

The paper proposes an end-to-end forensic pipeline using steganographic attribution and multimodal harm detection to reliably trace and attribute harmful misuse of AI-generated imagery on social platforms.

This paper investigates a novel security vulnerability where imperceptible branding hints can be injected into images and subsequently re-rendered onto new objects by generative AI models, proposing both attack scenarios and a robust mitigation solution.