publications

publications by categories in reversed chronological order. generated by jekyll-scholar.

2026

  1. ICLR
    Larger Datasets Can Be Repeated More: A Theoretical Analysis of Multi-Epoch Scaling in Linear Regression
    Tingkai Yan*, Haodong Wen*, Binghui Li*, and 3 more authors
    (*: equal contribution) , 2026
    arXiv
  2. ICLR
    How Learning Rate Decay Wastes Your Best Data in Curriculum-Based LLM Pretraining
    Kairong Luo, Zhenbo Sun, Haodong Wen, and 5 more authors
    2026
    arXiv

2025

  1. NeurIPS
    Adam Reduces a Unique Form of Sharpness: Theoretical Insights Near the Minimizer Manifold
    Xinghan Li&, Haodong Wen&, and Kaifeng Lyu
    (&: equal contribution; alphabet ordering) , 2025
    arXiv
  2. ICLR
    A multi-power law for loss curve prediction across learning rate schedules
    Kairong Luo, Haodong Wen, Shengding Hu, and 5 more authors
    2025
    arXiv
  3. ICASSP
    Exploring the robustness of in-context learning with noisy labels
    Chen Cheng*, Xinzhi Yu*, Haodong Wen*, and 4 more authors
    (*: equal contribution) , 2025
    arXiv