Received the Rackham Conference Travel Grant
Zijian Huang
I am a Ph.D. student in Industrial and Operations Engineering at the University of Michigan, advised by Prof. Raed Al Kontar.
Prior to that, I received my undergraduate education in Industrial and Operations Engineering at the University of Michigan and Electrical and Computer Engineering at Shanghai Jiao Tong University.
I am interested in developing reliable and efficient AI systems through machine learning, statistics, and optimization. My current interests include Bayesian optimization, the evaluation and reliability of LLM-based agents and multi-agent systems, and sequential decision-making under limited computational budgets.
Last updated: October 2026
News
Received the Accenture IOE Scholarship at UMich.
Awarded National Scholarship at SJTU.
Selected Publications
Understanding Generalization in Diffusion Models via Probability Flow Distance
ICML Workshop on High-dimensional Learning Dynamics (HiLD), 2025.
We propose Probability Flow Distance (PFD), a theoretically grounded and computationally efficient metric to quantify distributional generalization in diffusion models. Using a teacher-student evaluation protocol, we empirically uncover several novel generalization behaviors.
