Gelei Deng

AudioTrust: Benchmarking The Multifaceted Trustworthiness of Audio Large Language Models

Turning Bias into Bugs: Bandit-Guided Style Manipulation Attacks on LLM Judges

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models

TREANT: Red-teaming Text-to-Image Models with Tree-based Semantic Transformations

Fine-Grained Verifiers: Preference Modeling as Next-token Prediction in Vision-Language Alignment

RSafe: Incentivizing proactive reasoning to build robust and adaptive LLM safeguards