publications

Papers and preprints on model quantization, diffusion models, and efficient machine learning.

* indicates equal contribution.

2026

  1. arXiv
    JustQuant: You Don’t Need Smoothing, SVD, or Rotation for 4-Bit Activation Quantization
    Kaicheng Yang*, Kaisen Yang*, Chunyu Liu*, Xianglong Yan, Haotong Qin, and 6 more authors
    arXiv preprint arXiv:2609.33601, 2026
    arXiv PDF Blog Website
  2. NeurIPS
    MLS-Bench: A Holistic and Rigorous Assessment of AI Systems on Building Better AI
    In NeurIPS (Oral), 2026
    arXiv PDF Website
  3. ICML
    RobuQ: Pushing DiTs to W1.58A2 via Robust Activation Quantization
    Kaicheng Yang*, Xun Zhang*, Haotong Qin, Yucheng Lin, Kaisen Yang, and 1 more author
    In International Conference on Machine Learning, 2026
    arXiv PDF Code
  4. ICML
    Q-DiT4SR: Exploration of Detail-Preserving Diffusion Transformer Quantization for Real-World Image Super-Resolution
    Xun Zhang*, Kaicheng Yang*, Hongliang Lu, Haotong Qin, Yong Guo, and 1 more author
    In International Conference on Machine Learning, 2026
    arXiv PDF Code
  5. ICML
    Improving Sampling for Masked Diffusion Models via Information Gain
    Kaisen Yang, Jayden Teoh, Kaicheng Yang, Yitong Zhang, and Alex Lamb
    In International Conference on Machine Learning, 2026
    arXiv PDF Code
  6. ICLR
    PT^2-LLM: Post-Training Ternarization for Large Language Models
    Xianglong Yan*, Chengzhu Bao*, Zhiteng Li, Tianao Zhang, Kaicheng Yang, and 4 more authors
    In International Conference on Learning Representations, 2026
    arXiv PDF Code

2025

  1. ICML
    BiMaCoSR: Binary One-Step Diffusion Model Leveraging Flexible Matrix Compression for Real Super-Resolution
    Kai Liu*, Kaicheng Yang*, Zheng Chen, Zhiteng Li, Yong Guo, and 3 more authors
    In International Conference on Machine Learning, 2025
    arXiv PDF Code
  2. arXiv
    TreeQ: Pushing the Quantization Boundary of Diffusion Transformer via Tree-Structured Mixed-Precision Search
    Kaicheng Yang, Kaisen Yang, Baiting Wu, Xun Zhang, Qianrui Yang, and 3 more authors
    arXiv preprint arXiv:2512.06353, 2025
    arXiv PDF Code