Hi,#

I am a third-year undergraduate student at Rutgers University. I am fortunate to collaborate with Prof.Ruixiang Tang throughout my undergraduate studies.

Research#

My research interests lie in the domain of Generative AI Safety Alignment. Ultimately, I hope to contribute to the development of AI Systems that are Harmless (able to effectively refuse harmful user requests), Stable (avoiding unnecessary refusals of benign inputs), and Powerful (maintaining their capabilities even after safety alignment).

At present, I am focused on LLM Safety Alignment, aiming to identify and improve upon the limitations of current safety alignment strategies in Large Language Models. My work is driven by the belief that robustness and alignment should not come at the cost of performance.

Looking ahead, I also plan to explore safety and alignment issues in multimodal models, including MLLMs, Text-to-Image (T2I), and Text-to-Video (T2V) systems. As these models gain broader real-world deployment, understanding and mitigating their unique safety risks will be critical.

Others#

I graduated from Yuexi High School, a place that holds a special meaning to me. I spent the first 18 years of my life in the beautiful county of Yuexi. If you are graduating from, or have graduated from Yuexi High School and are seeking advice or simply someone to talk to, I’d be more than happy to connect and offer any help I can.

I’m also passionate about discussing the development and support of students from resource-constrained regions. I believe this is an incredibly important topic—ensuring that students in these areas have access to quality education and growth opportunities is essential for promoting long-term equity and social progress. I’m always open to meaningful conversations and collaborative efforts on this front.