Qiong Zhang (张琼)

Division of Computing and Mathematical Sciences
Mohamed bin Zayed University of Artificial Intelligence (MBZUAI)

qiong.zhang@mbzuai.ac.ae GitHub Google Scholar CV

About

Hi! Welcome to my website. I’m Qiong (pronounced similarly to “Chiong” if that helps!).

I am currently an assistant professor in the Division of Computing and Mathematical Sciences at Mohamed bin Zayed University of Artificial Intelligence (MBZUAI). As a statistician by training, I have ventured into the world of AI, exploring how data and algorithms can unravel complex stories. My research bridges statistics and machine learning, with a focus on mixture reduction, empirical likelihood, and tabular foundation models for in-context learning. See a detailed description of my current research directions here.

I am passionate about addressing real-world problems through statistical and AI-driven approaches, and I believe that data is the most compelling storyteller of all. If you’d like to chat about statistics, AI, or the hidden narratives within your dataset, I’d be delighted to connect!

→ More about me (education, research interests)

News

Oct 2026 other I joined MBZUAI as an assistant professor in the Division of Computing and Mathematical Sciences.
Sep 2026 paper Architecture alignment with sparse priors in tabular foundation models — New preprint.
Sep 2026 paper TabClustPFN: A prior-data fitted network for tabular data clustering — Paper accepted at NeurIPS 2026.
Sep 2026 paper Byzantine-tolerant distributed learning of finite mixture models under partial corruptions — New preprint.
Jul 2026 paper FedSPM: Routing-enabled federated learning under dual heterogeneity via semiparametric mixture — New preprint.
May 2026 paper PFN-TS: Thompson Sampling for contextual bandits via prior-data fitted networks — New preprint.
Apr 2026 paper Forgettable federated linear learning with certified data removal — Paper accepted at IEEE TNNLS.
Mar 2026 paper Byzantine-tolerant distributed learning of finite mixture models — Paper accepted at JRSSB.
Mar 2026 paper Neyman-Pearson multiclass classification under label noise via empirical likelihood — New preprint.
Jan 2026 paper TabClustPFN: A prior-data fitted network for tabular data clustering — New preprint.
Jan 2026 paper Beyond aggregation: Guiding clients in heterogeneous federated learning — Paper accepted at ICLR 2026.
Sep 2025 paper Gaussian Herding across Pens: 3DGS compaction via Gaussian mixture reduction — Paper accepted at NeurIPS 2025 (spotlight).
May 2025 paper TabPFN: One model to rule them all? — New preprint on tabular foundation models for statistical estimation.

Selected Publications

Byzantine-tolerant distributed learning of finite mixture models

Qiong Zhang^*, Yan Shuo Tan^*, Jiahua Chen

Journal of the Royal Statistical Society, Series B, 2026
Gaussian mixture reduction with composite transportation divergence

Qiong Zhang, Archer Gong Zhang, Jiahua Chen

IEEE Transactions on Information Theory, 2024
Distributed learning of finite mixture models

Qiong Zhang, Jiahua Chen

Journal of Machine Learning Research, 2022
Beyond aggregation: Guiding clients in heterogeneous federated learning

Zijian Wang^🎓, Xiaofei Zhang, Xin Zhang, Yukun Liu, Qiong Zhang^📧

International Conference on Learning Representations (ICLR), 2026
Gaussian Herding across Pens: An optimal transport perspective on global Gaussian reduction for 3DGS Spotlight

Tao Wang^*🎓, Mengyu Li^*🎓, Geduo Zeng^🎓, Cheng Meng^📧, Qiong Zhang^📧

Neural Information Processing Systems (NeurIPS), 2025
TabPFN: One model to rule them all?

Qiong Zhang^*, Yan Shuo Tan^*, Qinglong Tian^*, Pengfei Li

Minor revision at JASA

→ Full publication list