Mengxiao Gao | School of Operations Research and Information Engineering, Cornell University

About Me

I am a second-year PhD student in the School of Operations Research and Information Engineering (ORIE) at Cornell University. I am broadly interested in understanding when modern generative learning is statistically stable and efficient. Previously, I worked on causal discovery algorithms and non-parametric statistical tests.

Before my PhD, I received my B.S. from Weiyang College, Tsinghua University, where I pursued a dual major in Mathematics & Physics and Civil Engineering & Systems.

I am always happy to chat — feel free to reach out via email!

Research

My current research interests lie in the statistical foundations of generative models. In particular, I am interested in:

Statistical properties of score estimation and diffusion models;
Learning dynamics with machine-generated data and model collapse.

Previously, I worked on general casual discovery algorithms. [see [2] [1]]

Publications & Preprints

[2] MixCIT: A Kernel Based Local-Polynomial Debiased Test for Conditional Independence on Mixed-Type Data
Mengxiao Gao, Kyra Gan, Promit Ghosal
[arXiv]

[1] Hybrid Top-Down Global Causal Discovery with Local Search for Linear and Nonlinear Additive Noise Models
Sujai Hiremath, Jacqueline R.M.A. Maasch, Mengxiao Gao, Promit Ghosal, Kyra Gan
[arXiv][NeurIPS 2024]

Education

Cornell University
2025.08 – Present
Ph.D. in Operations Research and Information Engineering

Tsinghua University
2021.09 – 2025.06
B.S. in Mathematics and Physics; Double Major in Civil Engineering and Systems

Miscellaneous

My favorite figure skating programs include:
Weijing Sui/Cong Han: Rain in Your Black Eyes
Alena Kostornaia: Twilight
Rika Kihira: The Fire Within