Chaowei Xiao

Chaowei Xiao is an Assistant Professor at the University of Wisconsin-Madison (starting 2023), affiliated with the School of Computer, Data & Information Sciences. His research focuses on securing AI systems, particularly exploring robustness in trustworthy machine learning, autonomous systems, and large language models (LLMs). He holds a Ph.D. from the University of Michigan, Ann Arbor, and a B.S. from Tsinghua University.

Before joining UW-Madison, he worked as a research scientist at NVIDIA (2020–2022) and at Arizona State University (2022–2023). His work bridges model and system perspectives to ensure practical robustness and provable guarantees in AI applications like autonomous driving, healthcare, and IoT.

Key research areas include adversarial robustness, AI security, and ethical AI. Recent contributions include frameworks for detecting LLM hallucinations, mitigating jailbreak attacks, and securing multi-modal systems. His work on diffusion models for adversarial purification and physical-world attacks on autonomous driving systems has been widely recognized.

Notable awards include the 2024 USENIX Security Distinguished Paper Award, ACM Gordon Bell Finalist (2024), and Schmidt Sciences AI2050 Fellowship. He has advised students like Xiaogeng Liu (NVIDIA Fellow) and secured grants from Amazon, Apple, and UW-Madison. His lab actively publishes at top venues like NeurIPS, ICML, and CVPR.