Guanyang Wang Assistant Professor Department of Statistics Rutgers University, New Brunswick Contact Office 453, Hill Center 110 Frelinghuysen Road Piscataway, NJ 08854 Email: guanyang.wang [at] rutgers [dot] edu

About me

Welcome! I am an assistant professor in the Department of Statistics at Rutgers University, New Brunswick.

I received my Ph.D. in Mathematics (Ph.D. minor in Statistics) from Stanford University in August 2020, advised by Prof. Persi Diaconis. Prior to that, I received my B.S. with Honors in Mathematics from University of Science and Technology of China (USTC) in Jun 2015.

With several friends and colleagues, we organize a weekly online seminar on Monte Carlo methods. Check out our recorded videos here!

My Erdős number is 2 ( [1] , [2] , and [3]. ) My research is partially supported by NSF DMS-2210849, NSF FET-2403007, and an Adobe Data Science Research Award. Thank you!

I serve as an Associate Editor for Mathematics of Operations Research and as an Area Chair for NeurIPS 2026.

People

I’m fortunate to work with such talented students.

• Budhaditya Halder (Statistics, sixth year Ph.D. student)

• Jing Jia (Computer Science, fourth year Ph.D. student, co-advised with Peng Zhang)

Alumni

• Wei Yuan (Statistics Ph.D., graduated in 2026, First Job: Senior Computer Vision Engineer at 荣耀 (HONOR))

• Yasa Syed (Statistics Ph.D., graduated in 2025, First Job: Quantitative Researcher at Venerable)

Research Interests

My research primarily focuses on Monte Carlo methods, generative AI, quantum computing, and probability.

If you'd like to read some of my recent articles, you may want to look at:

Recent News

• New blog post: LLMs for Proof Generation and Verification [中文版] (Jul, 2026).

• New paper Spectral Gap for the Binary Fixed-Margin Swap Chain with Weibo Fu and Qian Qin resolves the 1997 KTV conjecture for binary matrices with an AI-generated, Lean-verified proof. [Blog] [Lean] (Jun, 2026).

• New paper Couple to Control: Joint Initial Noise Design in Diffusion Models with Jing Jia and Liyue Shen is available on arXiv (May, 2026).

• Jing Jia and Wei Yuan's first-author paper Weak Diffusion Priors Can Still Achieve Strong Inverse-Problem Performance with Sifan Liu, and Liyue Shen is accepted at ICML 2026 as a spotlight (top 2.2%). Check out the 机器之心 coverage! (Apr, 2026) 🎉

• Wei Yuan's first-author paper Markov chain Monte Carlo without evaluating the target: an auxiliary variable approach is accepted at ICML 2026 as an oral (top 0.7%) (Apr, 2026) 🎉.

• Paper Optimal Quantum Speedups for Repeatedly Nested Expectation Estimation with Yihang Sun and Jose Blanchet is accepted at ICML 2026 (Apr, 2026).

• Happy to receive the Presidential Fellowship for Teaching Excellence from Rutgers University. Thank you! (Apr, 2026)

• Serving as an Area Chair at NeurIPS 2026. (Apr, 2026)

• Co-organizing the Quantum Metrics and Performance Evaluation Workshop (QMETRICS 2026), a SIGMETRICS 2026 workshop. (Mar, 2026)

• Jing 's first lead-author paper Antithetic noise in diffusion models is accepted at ICLR 2026 🎉 (Jan, 2026)

• New paper TabClustPFN: A Prior-Fitted Network for Tabular Data Clustering with Tianqi Zhao , Yan Shuo Tan, and Qiong Zhang is on arxiv. (Jan, 2026)

• Happy to receive a teaching grant from Thinking Machines Lab for my Stochastic Processes (Stat 654). Thank you! (Dec, 2025)

• Happy to join the editorial board of Mathematics of Operations Research as an associate editor. (Dec, 2025)

• With Jose Blanchet and Kimi Sun , we delievred a tutorial on quantum computing at Winter Simulation Conference (WSC) 2025 . (Dec, 2025)

• Paper A phase transition in sampling from Restricted Boltzmann Machines with Youngwoo Kwon, Qian Qin, and Yuchen Wei is accepted at the Annals of Applied Probability (AoAP) (Oct, 2025).

• Paper Quantum speedup of non-linear Monte Carlo problems with Jose Blanchet, Yassine Hamoudi and Mario Szegedy is accepted at NeurIPS 2025 as a spotlight (Sep, 2025).

• Paper CCS: Controllable and Constrained Sampling with Diffusion Models via Initial Noise Perturbation with Bowen Song, Liyue Shen and others is accepted at NeurIPS 2025 (Sep, 2025).

Recent Works

• Spectral Gap for the Binary Fixed-Margin Swap Chain
  Weibo Fu, Qian Qin, Guanyang Wang^* [arXiv] [Code] [Blog]
  arXiv, 2026

• Couple to Control: Joint Initial Noise Design in Diffusion Models
  Jing Jia, Liyue Shen, Guanyang Wang [arxiv]
  arXiv, 2026

• Weak Diffusion Priors Can Still Achieve Strong Inverse-Problem Performance
  Jing Jia ^†⟐ , Wei Yuan ^†, Sifan Liu, Liyue Shen, Guanyang Wang^⟐ [arxiv] [Page] [Code] [机器之心] [知乎]
  ICML 2026, spotlight , top 2.2%

• Markov chain Monte Carlo without evaluating the target: an auxiliary variable approach
  Wei Yuan ^**, Guanyang Wang^⟐ [arXiv] [Code]
  ICML 2026, oral , top 0.7%

• Optimal Quantum Speedups for Repeatedly Nested Expectation Estimation
  Yihang Sun, Guanyang Wang, Jose Blanchet [arXiv]
  ICML 2026

• TabClustPFN: A Prior-Fitted Network for Tabular Data Clustering
  Tianqi Zhao, Guanyang Wang , Yan Shuo Tan, Qiong Zhang [arxiv]

• Antithetic Noise in Diffusion Models
  Jing Jia ^**, Sifan Liu ^⟐, Bowen Song, Wei Yuan, Liyue Shen ^⟐, Guanyang Wang^⟐ [arxiv] [Page] [Blog] [Code] [知乎]
  ICLR 2026

• CCS: Controllable and Constrained Sampling with Diffusion Models via Initial Noise Perturbation
  Bowen Song, Zecheng Zhang, Zhaoxu Luo, Jason Hu, Wei Yuan, Jing Jia, Zhengxu Tang , Guanyang Wang^⟐, Liyue Shen ^⟐ [arXiv]
  NeurIPS 2025

• Quantum speedup of non-linear Monte Carlo problems
  Jose Blanchet, Yassine Hamoudi, Mario Szegedy, Guanyang Wang^* [arXiv]
  NeurIPS 2025, spotlight

• Connecting Quantum Computing with Classical Stochastic Simulation
  Jose Blanchet, Mark S. Squillante, Mario Szegedy, Guanyang Wang^*
  2025 Winter Simulation Conference (WSC) , accepted

• On importance sampling and independent Metropolis-Hastings with an unbounded weight function
  George Deligiannidis, Pierre E. Jacob, El Mahdi Khribch , Guanyang Wang^* [arXiv]
  Major revision, the Annals of Statistics

• A phase transition in sampling from Restricted Boltzmann Machines
  Youngwoo Kwon, Qian Qin, Guanyang Wang^*, Yuchen Wei [arXiv]
  the Annals of Applied Probability (AoAP) , 36(2): 1615-1652

* = alphabetical order, † = equal contribution, ⟐ = corresponding author, ** = student first author

Miscellaneous

• Chinese Painting Generation Using Generative Adversarial Networks
  Guanyang Wang, Ying Chen, Yuan Chen
  CS231n Course Project, 2017. [Chinese Painting Dataset]

• Pattern detection in bipartite networks: a review of terminology, applications and methods
  Zachary Neal, Annabel Cadieux, Diego Garlaschelli, Nicholas J. Gotelli, Fabio Saracco, Tiziano Squartini, Shade T. Shutters, Werner Ulrich, Guanyang Wang, Giovanni Strona
  Accepted, PLOS Complex Systems . [arXiv]