Yichi Zhang

Towards Safe Reasoning in Large Reasoning Models via Corrective Intervention

Unveiling the Basin-Like Loss Landscape in Large Language Models

Towards the Worst-case Robustness of Large Language Models

STAIR: Improving Safety Alignment with Introspective Reasoning

Mitigating Overthinking in Large Reasoning Models via Manifold Steering

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

Towards Mitigating Factual Hallucination in LLMs through Self-Alignment with Memory