Publications

(*) denotes equal contribution.

2025

  1. NeurIPS 2025
    Efficient Multimodal Dataset Distillation via Generative Models
    Jiaqi Xue, Mayank Kumar, Yuzhang Shang, Shangqian Gao, Mengxin Zheng, Xiaoqian Jiang, and Qian Lou
    Neural Information Processing Systems (NeurIPS) 2025
  2. NeurIPS 2025
    DictPFL: Efficient and Private Federated Learning on Encrypted Gradients
    Zhenghao Zhao, Haoxuan Wang, Junyi Wu, Yuzhang Shang, Gaowen Liu, and Yan Yan
    Neural Information Processing Systems (NeurIPS) 2025
  3. ICCV 2025
    CaO2: Rectifying Inconsistencies in Diffusion-Based Dataset Distillation
    Haoxuan Wang, Zhenghao Zhao, Junyi Wu, Yuzhang Shang, Gaowen Liu, and Yan Yan
    International Conference on Computer Vision (ICCV) 2025
  4. ICCV 2025
    DLFR-Gen: Diffusion-based Video Generation with Dynamic Latent Frame Rate
    Zhihang Yuan, Rui Xie, Yuzhang Shang, Hanling Zhang, Siyuan Wang, Shengen Yan, Guohao Dai, and 1 more author
    International Conference on Computer Vision (ICCV) 2025
  5. ICCV 2025
    QuEST: Low-bit Diffusion Model Quantization via Efficient Selective Finetuning
    Haoxuan Wang, Yuzhang Shang, Zhihang Yuan, Junyi Wu, Junchi Yan, and Yan Yan
    International Conference on Computer Vision (ICCV) 2025
    arXiv Code
  6. ICCV 2025
    Robin3d: Improving 3d large language model via robust instruction tuning
    Weitai Kang, Haifeng Huang, Yuzhang Shang, Mubarak Shah, and Yan Yan
    International Conference on Computer Vision (ICCV) 2025
  7. ACMMM 2025
    Dlfr-vae: Dynamic latent frame rate vae for video generation
    Zhihang Yuan, Siyuan Wang, Rui Xie, Hanling Zhang, Tongcheng Fang, Yuzhang Shang, Shengen Yan, and 2 more authors
    ACM Multimedia 2025
  8. ACL 2025
    GSQ-Tuning: Group-Shared Exponents Integer in Fully Quantized Training for LLMs On-Device Fine-tuning
    Sifan Zhou, Shuo Wang, Zhihang Yuan, Mingjia Shi, Yuzhang Shang, and Dawei Yang
    Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics (ACL) 2025
  9. ACL 2025
    PTQ1. 61: Push the Real Limit of Extremely Low-Bit Post-Training Quantization Methods for LLMs
    Jiaqi Zhao, Miao Zhang, Ming Wang, Yuzhang Shang, Zhang Kaihao, Weili Guan, Yaowei Wang, and 1 more author
    Findings of the 48th Annual Meeting of the Association for Computational Linguistics (ACL) 2025
  10. CVPR 2025
    DKDM: Data-Free Knowledge Distillation for Diffusion Models with Any Architecture
    Qianlong Xiang, Miao Zhang, Yuzhang Shang, Jianlong Wu, Yan Yan, and Liqiang Nie
    Computer Vision and Pattern Recognition (CVPR) 2025
  11. CVPR 2025
    Distilling Long-tailed Datasets
    Zhenghao Zhao, Haoxuan Wang, Yuzhang Shang, Kai Wang, and Yan Yan
    Computer Vision and Pattern Recognition (CVPR) 2025
  12. CVPR 2025
    A closer look at time steps is worthy of triple speed-up for diffusion model training
    Kai Wang, Mingjia Shi, Yukun Zhou, Zekai Li, Zhihang Yuan, Yuzhang Shang, Xiaojiang Peng, and 2 more authors
    Computer Vision and Pattern Recognition (CVPR) 2025
  13. LLaVA-PruMerge: Adaptive Token Reduction for Efficient Large Multimodal Models
    Yuzhang Shang, Mu Cai, Bingxin Xu, Yong Jae Lee, and Yan Yan
    International Conference on Computer Vision (ICCV) 2025
    arXiv Code
    The first token reduction method for accelerating Multimodal LLM.

2024

  1. NeurIPS 2024
    PTQ4DiT: Post-training Quantization for Diffusion Transformers
    Junyi Wu, Haoxuan Wang, Yuzhang Shang, Mubarak Shah, and Yan Yan
    Conference on Neural Information Processing Systems (NeurIPS) 2024
  2. NeurIPS 2024
    HEPrune: Fast Private Training of Deep Neural Networks With Encrypted Data Pruning
    Yancheng Zhang, Mengxin Zheng, Yuzhang Shang, Xun Chen, and Qian Lou
    Conference on Neural Information Processing Systems (NeurIPS) 2024
  3. ECCV 2024
    Dataset Quantization with Active Learning-based Adaptive Sampling
    Zhenghao Zhao, Yuzhang Shang, Junyi Wu, and Yan Yan
    European Conference on Computer Vision (ECCV) 2024
  4. PB-LLM: Partially Binarized Large Language Models
    Yuzhang Shang, Zhihang Yuan, and Zhen Dong
    International Conference on Learning Representations (ICLR) 2024
    arXiv Code
    The first binarization exploration for LLMs.
  5. E-CAR: Efficient Continuous Autoregressive Image Generation via Multistage Modeling
    Zhihang Yuan*, Yuzhang Shang*, Hanling Zhang, Tongcheng Fang, Rui Xie, Bingxin Xu, Yan Yan, and 3 more authors
    arXiv Dec. 2024
    arXiv
  6. LLM Inference Unveiled: Survey and Roofline Model Insights
    Zhihang Yuan*, Yuzhang Shang*, Yang Zhou*, Zhen Dong, Chenhao Xue, Bingzhe Wu, Zhikai Li, and 6 more authors
    arXiv Feb. (Survey Paper) 2024
    arXiv Code
  7. CVPR 2024
    Enhancing Post-training Quantization Calibration through Contrastive Learning
    Yuzhang Shang, Gaowen Liu, Ramana Kompella, and Yan Yan
    Computer Vision and Pattern Recognition (CVPR) 2024
  8. CVPR 2024
    Efficient Multitask Dense Predictor via Binarization
    Yuzhang Shang, Dan Xu, Gaowen Liu, Ramana Kompella, and Yan Yan
    Computer Vision and Pattern Recognition (CVPR) 2024
    arXiv Code
  9. ICRA 2023
    FBPT: A Fully Binary Point Transformer
    Zhixing Hou, Yuzhang Shang, and Yan Yan
    International Conference on Robotics and Automation (ICRA) 2024
    arXiv

2023

  1. PTQ4DM: Post-training Quantization on Diffusion Models
    Yuzhang Shang, Zhihang Yuan, Bin Xie, Bingzhe Wu, and Yan Yan
    Computer Vision and Pattern Recognition (CVPR) 2023
    arXiv Code
    The first network compression method for diffusion models.
  2. ASVD: Activation-aware Singular Value Decomposition for Compressing Large Language Models
    Zhihang Yuan*, Yuzhang Shang*, Yue Song, Qiang Wu, Yan Yan, and Guangyu Sun
    arXiv Dec. 2023
    arXiv Code
    The first low-rank decomposition method for LLMs. The concept of low-rank attention was later adopted in DeepSeek-v2 (six months afterward).
  3. NeurIPS 2023
    MIM4DD: Mutual Information Maximization for Dataset Distillation
    Yuzhang Shang, Zhihang Yuan, and Yan Yan
    Conference on Neural Information Processing Systems (NeurIPS) 2023
    arXiv
  4. ICCV 2023
    Causal-DFQ: Causality Guided Data-free Network Quantization
    Yuzhang Shang, Bingxin Xu, Gaowen Liu, Ramana Kompella, and Yan Yan
    International Conference on Computer Vision (ICCV) 2023
    arXiv Code
  5. arXiv 2023
    RPTQ: Reorder-based Post-training Quantization for Large Language Models
    Zhihang Yuan, Lin Niu, Jiawei Liu, Wenyu Liu, Xinggang Wang, Yuzhang Shang, Guangyu Sun, and 3 more authors
    arXiv 2023
    arXiv Code

2022

  1. ECCV 2022
    Network Binarization via Contrastive Learning
    Yuzhang Shang, Dan Xu, Ziliang Zong, Liqiang Nie, and Yan Yan
    European Conference on Computer Vision (ECCV) 2022
    arXiv Code
  2. ECCV 2022
    Lipschitz Continuity Retained Binary Neural Network
    Yuzhang Shang, Dan Xu, Bin Duan, Ziliang Zong, Liqiang Nie, and Yan Yan
    European Conference on Computer Vision (ECCV) 2022
    arXiv Code
  3. ICASSP 2022
    Win The Lottery Ticket Via Fourier Analysis: Frequencies Guided Network Pruning
    Yuzhang Shang, Bin Duan, Ziliang Zong, Liqiang Nie, and Yan Yan
    IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2022
    arXiv

2021

  1. ICCV 2021
    Lipschitz Continuity Guided Knowledge Distillation
    Yuzhang Shang, Bin Duan, Ziliang Zong, Liqiang Nie, and Yan Yan
    International Conference on Computer Vision (ICCV) 2021
    arXiv Code