Kai-Wei Chang

Kai-Wei Chang is an assistant professor at UCLA in the Computer Science Department, working on natural language processing, machine learning, and AI safety. His research focuses on building robust and interpretable NLP systems, including work on adversarial robustness and structured prediction.

Sources in this wiki

• FLIRT: Feedback Loop In-context Red Teaming

Topics

• NLP, AI Safety, Model safety

Notes

Co-author on FLIRT; known for research on adversarial robustness in NLP and structured prediction.