About Me


I am an associate professor in Computer Science and Data Science at UChicago, the director of the SIGMA Lab (Strategic IntelliGence for Machine Agents) and an AI2050 early career fellow. Currently, I am on leave from UChicago and work full time at Google to help train LLMs.



Contact


Office: Room 312, 5460 S. University Ave

E-mail: haifengxu AT uchicago DOT edu


About My Research


My research broadly centers on AI agents, spanning their theoretical foundations, capability advancement, multi-agent interactions, and real-world applications. Recently, I have become particularly interested in developing AI agents that can accurately forecast future events -- a problem of auto-research nature but differs intrinsically from other STEM research tasks. Needless to mention, strong forecasting has immense practical value. Some of our lab's example projects include building the Prophet Arena platform and exploring how AI forecasters can be leveraged for profitable betting in prediction markets. More information can be found at a recent ICML-26 workshop and a a fun forecasting hackathon our lab hosted.

In the past, I also studied economic foundations of AI agents (see Schmidt Sciences' interview and my talk at Sigecom Winter25 meeting), intelligent communications among agents (see a tutorial organized by my student at WINE24), and incentive-aligned machine learning (see my talk at Princeton).

Recent News

PHP Hits Count