Yuqi Gu

Email: yuqi.gu@columbia.edu.
Address: Room 928 SSW, 1255 Amsterdam Avenue, New York, NY 10027

I am an Assistant Professor in the Department of Statistics at Columbia University. I am also a member of the Data Science Institute. Before joining Columbia in 2021, I spent a year as a postdoc at Duke University, mentored by David B. Dunson. In 2020 I received a Ph.D. in Statistics from the University of Michigan, advised by Gongjun Xu. In 2015 I received a B.S. in Mathematics from Tsinghua University. My first name can be pronounced as /ju:-tʃi:/. My name in Chinese is 顾雨琦.

My research develops statistical theory and methods for uncovering latent structure in modern complex data. A unifying theme is to make latent structure and representation learning identifiable, interpretable, computationally scalable, and statistically reliable.

  • Identifiable deep generative models and causal representation learning: I study identifiability, latent graph discovery, and causal representation learning in nonlinear probabilistic graphical models with latent structures.
  • High-dimensional statistical inference for latent structure: The high dimensionality and latent structure pose double statistical challenges. I develop spectral, tensor, and likelihood-based methods for mixture, mixed-membership, and nonlinear low-rank representation problems, with finite-sample theory and uncertainty quantification.
  • Latent variable models for psychometrics, heterogeneous data, and AI evaluation: I propose principled latent variable models for educational, psychological, biomedical, and language-model data, including cognitive diagnosis, item response theory, and psychometric frameworks for evaluating large language models (LLMs).

Here is my CV.

Representative Publications (see full list)

Underlined are student or postdoc authors under my supervision. ✉ indicates I am the corresponding author.

    • Deep Discrete Encoders: Identifiable Deep Generative Models for Rich Data with Discrete Latent Layers
      Seunghyun Lee, and Yuqi Gu✉
      Journal of the American Statistical Association (2026), 121 (553): 194–208. [arXiv] [Journal] [Code]
    • Generalized Grade-of-Membership Estimation for High-dimensional Locally Dependent Data
      Ling Chen*, Chengzhu Huang*, and Yuqi Gu✉
      Journal of the American Statistical Association (2026), accepted. [arXiv] [Journal] [Code]
    • Adaptive Transfer Clustering: A Unified Framework
      Yuqi Gu⁺, Zhongyuan Lyu⁺, and Kaizheng Wang⁺
      Journal of the American Statistical Association (2026), accepted. [arXiv] [Journal] [Code]
    • Latency-Response Theory Model: Evaluating Large Language Models via Response Accuracy and Chain-of-Thought Length
      Zhiyu Xu, Jia Liu, Yixin Wang, and Yuqi Gu✉
      Annals of Applied Statistics (2026), accepted. [arXiv] [Code]
    • On Theoretical Identifiability of Discrete Latent Causal Graphical Models
      Seunghyun Lee, and Yuqi Gu
      Transactions on Machine Learning Research (2026) [arXiv]
    • Degree-heterogeneous Latent Class Analysis for High-dimensional Discrete Data
      Zhongyuan Lyu, Ling Chen, and Yuqi Gu✉
      Journal of the American Statistical Association (2025), 120 (552): 2435–2448. [arXiv] [Journal] [Code]
    • Blessing of Dependence: Identifiability and Geometry of Discrete Models with Multiple Binary Latent Variables
      Yuqi Gu✉
      Bernoulli (2025), 31 (2): 948–972. [arXiv] [Journal]
    • Bayesian Pyramids: Identifiable Multilayer Discrete Latent Structure Models for Discrete Data
      Yuqi Gu✉, and David B. Dunson
      Journal of the Royal Statistical Society Series B: Statistical Methodology (2023), 85 (2): 399–426. [arXiv] [Journal] [Code]
    • A Joint MLE Approach to Large-Scale Structured Latent Attribute Analysis
      Yuqi Gu✉, and Gongjun Xu
      Journal of the American Statistical Association (2023), 118 (541): 746–760. [arXiv] [Journal]
    • Partial Identifiability of Restricted Latent Class Models
      Yuqi Gu, and Gongjun Xu
      Annals of Statistics (2020), 48 (4): 2082–2107. [arXiv] [Journal]