Starting in Summer 2025, I am an Assistant Professor at the College of Engineering & Computer Science (CECS) of VinUniversity. My current research interests are on generative models and optimization, with a focus on methodological/theoretical side.
Before joining VinUniversity, from 2023-2025, I was a Research Fellow at Mathematics Department of National University of Singapore (NUS), working with Soh Yong Sheng. Before that, in 2022, I was a postdoctoral researcher at Télecom Paris, working with Florence d’Alché, Gabriel Peyré and Remi Flamary on the applications of optimal transport to machine learning.
I obtained my PhD in 2021 within the Proba-Stat team of Département de Mathématiques d’Orsay and INRIA Parietal team, working with problems in high-dimensional statistics. I was very fortunate to be advised by Sylvain Arlot and Bertrand Thirion.
I spent a fair amount of time during my PhD for the developments of open-source software, and still code daily (though less than I want to). Check out Nilearn, hidimstat, and Benchopt for the open-source projects that I have been developing/contributing to.
At Binh’s AI Lab (BAIL), we are doing a mix of theory and algorithmic development, with focus on generative modeling and optimization.
Members:
For the full list, see my Google Scholar. *: Equal contributions. †: Equal advising.
(LLM Alignment, Steering & Evaluation)
Phan, N., Huynh, T., Tran, KT., Nguyen, DA., Nguyen, PTA., Nguyen, T., Chawla, NV., Buntine, W., Wong, KS., Doan, KD., BN. PRISM: A Multi-Dimensional Benchmark for Evaluating LLM Peer Reviewers. Under review. [paper]
Nguyen, MP., La, CD., Nguyen, DMH., Chawla, NV., BN†, Doan, KD.† The Reasoning Boundary Paradox: How Reinforcement Learning Constrains Language Models. Under review (2025+). [paper]
Nguyen, NB., BN, Nguyen, DA., Nguyen, VA. Distributional Surgery for Language Model Activations. EMNLP 2025 Findings. [paper] [code]
Nguyen, TH., Nguyen, NB., BN, Nguyen, VA. Task-driven Layerwise Additive Activation Intervention. NAACL 2025.
Nam Nguyen, Ly Tran Ho Khanh, Thanh Nguyen-Tang, BN. Online Preference Optimization for Multi-Objective Alignment of Generative Models. Under review.
(Optimal Transport & Optimization)
Nguyen, CL., Nguyen, N., BN. Quadratically Regularized Optimal Transport: Localization Bounds and Affine Case Analysis. ICML 2026. [paper]
Nguyen, HA., BN, Soh, YS. Sum-of-Squares Hierarchy for the Gromov Wasserstein Problem. SIAM J. Optimization, 2026. [paper]
Le, B.*, Dao, T.*, BN†, Chu, H.† Tight Robustness Certificates and Wasserstein Distributional Attacks for Deep Neural Networks. Under review (2025+). [paper]
Chen, JY.*, BN*, Soh, YS. Semidefinite Relaxations of the Gromov-Wasserstein Distance. NeurIPS 2024. [paper] [code]
Moreau, T. et al. Benchopt: Reproducible, efficient and collaborative optimization benchmarks. NeurIPS 2022. [paper] [code]
(Generative & Diffusion Models)
Nguyen, NB., BN, Nguyen, TH., Nguyen, VA. Generative Conditional Distributions by Neural (Entropic) Optimal Transport. ICML 2024. [paper] [code]
Nguyen, NB., BN, Nguyen, VA. Bellman Optimal Step-size Straightening of Flow-Matching Models. ICLR 2024. [paper] [code]
Dao, Q., Phung, H., BN, & Tran, A. Flow Matching in Latent Space. arXiv 2023. [paper] [code]
(Statistical Inference)
BN, Thirion, B., & Arlot, S. A Conditional Randomization Test for Sparse Logistic Regression in High-Dimension. NeurIPS 2022. [paper] [code]
BN, J.-A. Chevalier, B. Thirion, & S. Arlot. Aggregation of Multiple Knockoffs. ICML 2020. [paper] [code]
tuanbinhs [AT] gmail.com
I am recruiting junior researchers at all ranks working at the CECS of VinUniversity:
The topics are flexible, but should be based on my research interests. Please contact me by email.