Research

Research Interests

$\ast$ Corresponding author; $\dagger$ Joint first author; $\ddagger$ Alphabetical order

Mengyu Li, Qianqian Qu, Jun S. Liu$^\ast$ (2026). BridgeTwist: Twisting Schrödinger bridges for training-free conditional sampling. NeurIPS 2026. [paper]
Mengyu Li, Meng Cheng$^\ast$, Jun S. Liu$^\ast$. Free-support Wasserstein resampling in sequential Monte Carlo. Submitted.
Xinyue Yang, Mengyu Li$^\ast$, Jun S. Liu$^\ast$. Generative distributional regression via joint and conditional distribution matching. Submitted.
Pengyu Li, Mengyu Li$^\dagger$, Jianzhu Ma, Jun S. Liu$^\ast$. Steer, select, and stratify: Training-free sampling for protein motif scaffolding. Submitted.

Mengyu Li, Cheng Meng$^\ast$, Xiaodan Fan (2026). Iterative optimal transport for multimodal image registration. Pattern Recognition. [paper] [code]
Mengyu Li, Bencong Zhu$^\dagger$, Cheng Meng$^\ast$, Xiaodan Fan$^\ast$ (2025). Double optimal transport for differential gene regulatory network inference with unpaired samples. Bioinformatics. [paper] [code]
Tao Wang, Mengyu Li$^\dagger$, Geduo Zeng, Cheng Meng$^\ast$, Qiong Zhang$^\ast$ (2025). Gaussian herding across pens: An optimal transport perspective on global Gaussian reduction for 3DGS. NeurIPS 2025 Spotlight (Top 3%). [paper]

Xiaxue Ouyang, Hao Zheng, Haoxian Liang, Jingyi Zhang, Yixuan Qiu, Cheng Meng, Mengyu Li$^\ast$ (2026). Sparsification techniques for large‐scale optimal transport problems. Wiley Interdisciplinary Reviews: Computational Statistics. [paper]
Yukuan Hu, Mengyu Li$^\ddagger$, Xin Liu, Cheng Meng (2025). Sampling-based methods for multi-block optimization problems over transport polytopes. Mathematics of Computation. [paper]
Mengyu Li, Jun Yu$^\dagger$, Tao Li, Cheng Meng$^\ast$ (2023). Importance sparsification for Sinkhorn algorithm. Journal of Machine Learning Research. (Student Paper Award in 2023 JSM, Statistics in Imaging Section). [paper] [code]
Mengyu Li, Jun Yu$^\dagger$, Hongteng Xu, Cheng Meng$^\ast$ (2023). Efficient approximation of Gromov-Wasserstein distance using importance sparsification. Journal of Computational and Graphical Statistics. [paper] [code]

Dunyao Xue, Mengyu Li$^\ast$, Cheng Meng, Jingyi Zhang (2026). Core-elements subsampling for alternating least squares. Journal of Computational and Graphical Statistics. [paper]
Junyi Lin, Mengyu Li$^\dagger$, Cheng Meng$^\ast$, Yongdao Zhou$^\ast$ (2026). Sparsification subsampling for partial least squares regression. Journal of Computational and Graphical Statistics. [paper]
Junlie Huang, Xinlai Kang, Qiannan Huang, Mengyu Li, Cheng Meng$^\ast$ (2025). Efficient approximation of leverage scores in two-dimensional autoregressive models with application to image anomaly detection. Journal of Computational and Graphical Statistics. [paper]
Mengyu Li, Jingyi Zhang, Cheng Meng$^\ast$ (2024). Nonparametric additive models for billion observations. Journal of Computational and Graphical Statistics. [paper] [code]
Mengyu Li, Jun Yu$^\dagger$, Tao Li, Cheng Meng$^\ast$ (2024). Core-elements for large-scale least squares estimation. Statistics and Computing. (Student Paper Award in 2022 JSM, Sections on Statistical Computing and Graphics). [paper] [code]

Xinlai Kang, Mengyu Li, Xuqiang Chen, Fangyu Li, Cheng Meng$^\ast$ (2023). A Hausdorff regression paradigm for interval privacy. IEEE Signal Processing Letters. [paper]
Mengyu Li, Junlong Zhao$^\ast$ (2022). Communication-efficient distributed linear discriminant analysis for binary classification. Statistica Sinica. [paper]