Yang Cao is an Associate Professor at the Institute of Science Tokyo and the head of the Trustworthy Data Science and AI (TDSAI) Lab. His research focuses on a central question: how to build AI systems that are both powerful and trustworthy. He studies the privacy, security, and reliability of modern AI and data systems, spanning large language models, federated learning, agentic AI systems, and data infrastructures (e.g., vector databases and data markets), aiming to establish rigorous foundations and practical systems that bridge machine learning, data systems, and security.
He received his Ph.D. from Kyoto University and has held positions at Emory University, Kyoto University, and Hokkaido University, as well as a visiting researcher role at Meta. His work has been published in leading venues and recognized by awards including the IEEE Computer Society Japan Young Author Award, the DBSJ Kambayashi Young Researcher Award, and multiple Best/Outstanding Paper Awards. He also serves the research community as an Associate Editor and Program Committee member for major journals and conferences such as VLDB, SIGMOD, KDD, ACM CCS and IEEE TDSC.
