Chenyu Zheng

Ph.D. Student at Renmin University of China

profile.jpg

I am a third-year Ph.D. student in the GSAI-ML Group, Renmin University of China, advised by Prof. Chongxuan Li. Currently, I am also a research intern at ByteDance Seed. Before that, I received my B.E. degree from the School of Computer Science, Wuhan University in 2023.

My primary interest is machine learning theory. Recently, I focus on optimization and scalability in deep learning.

News

Apr 27, 2026 Our work “Spectral Condition for μP under Width-Depth Scaling” received an Outstanding Paper Award at the ICLR Workshop on Deep Generative Models: Theory, Principle, and Efficacy.
Apr 09, 2026 I gave a tutorial “An Overview of Maximal Update Parametrization (μP)”. [Slides]
Jan 01, 2026 I was supported by the Young Scientists (Ph.D.) Fund from NSFC (国家自然科学基金博士生专项).

Selected publications

  1. ICLRW
    Spectral Condition for μP under Width-Depth Scaling
    Chenyu Zheng, Rongzhen Wang, Xinyu Zhang, and 1 more author
    ICLR Workshop on Deep Generative Models: Theory, Principle, and Efficacy, 2026
  2. NeurIPS
    Scaling Diffusion Transformers Efficiently via μP
    Chenyu Zheng, Xinyu Zhang, Rongzhen Wang, and 5 more authors
    In Advances in Neural Information Processing Systems, 2025
  3. NeurIPS
    On Mesa-Optimization in Autoregressively Trained Transformers: Emergence and Capability
    Chenyu Zheng, Wei Huang, Rongzhen Wang, and 3 more authors
    In Advances in Neural Information Processing Systems, 2024
  4. NeurIPS
    Toward Understanding Generative Data Augmentation
    Chenyu Zheng, Guoqiang Wu, and Chongxuan Li
    In Advances in Neural Information Processing Systems, 2023
  5. ICML
    Revisiting Discriminative vs. Generative Classifiers: Theory and Implications
    Chenyu Zheng, Guoqiang Wu, Fan Bao, and 3 more authors
    In International Conference on Machine Learning, 2023