Xingxing Wei

Towards Seed-Invariant Safety Alignment in Text-to-Image Models

Enhancing Adversarial Robustness of Vision Language Models via Adversarial Mixture Prompt Tuning

Strata-Sword: A Hierarchical Safety Evaluation towards LLMs based on Reasoning Complexity of Jailbreak Instructions

Mitigating Overthinking in Large Reasoning Models via Manifold Steering

TASAR: Transfer-based Attack on Skeletal Action Recognition

DEAL: High-Efficacy Privacy Attack on Retrieval-Augmented Generation Systems via LLM Optimizer

Adaptive Strategy Evolution for Generating Tailored Jailbreak Prompts against Black-Box Safety-Aligned LLMs