I am a third-year Ph.D. student in Deparment of Statistics & Data Science at University of California, Los Angeles(UCLA), where I am advised by Prof. Guang Cheng at Trustworthy AI Lab. My current research focuses on statistical evaluation of generative data fidelity. I also collaborated with Prof. Minjeong Jeon on developing effect estimations for transition diagnostic classification models. Prior to UCLA, I completed my bachelor’s degree of Applied Mathematics at Tongji University, advised by Prof. Li Zhang.

Feel free to reach out via email if you are interested in collaborating.

Research Topics

My current research focuses on developing effective methods for evaluating the fidelity of generative models—that is, assessing how well synthetic data replicates the characteristics of real data. High fidelity is essential for producing trustworthy outputs and realistic content in AI applications. Generative data fidelity encompasses multiple dimensions depending on the domain and task. Specifically, my current study involves:

1. Distributional fidelity in image generation
2. Causal fidelity in synthetic tabular data
3. Discriminative fidelity in language model–generated text

Education

• Ph.D in Statistics, University of California, Los Angeles
• M.S. in Statistics, University of California, Los Angeles, 2023
• B.S. in Applied Mathematics, Tongji University, 2021

Publications

1. Tao, L., Duan, H., Jeon, M. (2025).
   Bayesian Inference of Transition Diagnostic Classification Models with Multilevel Effects Using Gibbs Sampling with Pòlya-Gamma Augmentation.
   In progress.

2. Tao, L., Xu, S., Wang, C. H., Suh, N., Cheng, G. (2024).
   Discriminative Estimation of Total Variation Distance: A Fidelity Auditor for Generative Data.
   Preprint, 2024.
   arXiv:2405.15337

3. Chen, Y., Tao, L., Zhang, L. (2023).
   Injective Δ + 2 Coloring of Planar Graph Without Short Cycles.
   Acta Mathematicae Applicatae Sinica, English Series, 39(4), 1009–103.
   Springer Link