Xuandong Zhao's Team

My research lies at the intersection of Machine Learning (ML), Natural Language Processing (NLP), and AI Safety, with a focus on responsible and reliable generative AI. I aim to build solid theoretical foundations and practical algorithms that ensure the safety and security of large language models (LLMs) and AI agents. My work addresses the full lifecycle of responsible AI development, including: enhancing model alignment and robustness to mitigate risks; establishing the provenance of AI-generated content through watermarking; and protecting privacy, copyright, and intellectual property during model training and deployment.