Publications

2025

1. ICCV25 Highlight

   

   Rep-MTL: Unleashing the Power of Representation-level Task Saliency for Multi-Task Learning

   Zedong Wang, Siyuan Li, and Dan Xu

   IEEE/CVF International Conference on Computer Vision (ICCV), 2025

   Highlight (top 2.3%) Bib HTML PDF Code

   Rep-MTL was selected as Highlight presentation at ICCV 2025 on July 24, 2025.

   @inproceedings{wang2025repmtl,
     title = {Rep-MTL: Unleashing the Power of Representation-level Task Saliency for Multi-Task Learning},
     author = {Wang, Zedong and Li, Siyuan and Xu, Dan},
     booktitle = {IEEE/CVF International Conference on Computer Vision (ICCV)},
     year = {2025},
   }

2. 

   MergeVQ: A Unified Framework for Visual Generation and Representation with Disentangled Token Merging and Quantization

   Siyuan Li, Luyuan Zhang, Zedong Wang, and 8 more authors

   Conference on Computer Vision and Pattern Recognition (CVPR), 2025

   HF Daily Top-1 Bib PDF Code

   MergeVQ ranked top 1 at Hugging Face Daily Papers on Apr 3rd, 2025

   @inproceedings{li2025mergevq,
     title = {MergeVQ: A Unified Framework for Visual Generation and Representation with Disentangled Token Merging and Quantization},
     author = {Li, Siyuan and Zhang, Luyuan and Wang, Zedong and Tian, Juanxi and Tan, Cheng and Liu, Zicheng and Yu, Chang and Xie, Qingsong and Lu, Haonan and Wang, Haoqian and Lei, Zhen},
     booktitle = {Conference on Computer Vision and Pattern Recognition (CVPR)},
     year = {2025},
   }

3. 

   Taming LLMs by Scaling Learning Rates with Gradient Grouping

   Siyuan Li^*, Juanxi Tian^*, Zedong Wang^*, and 4 more authors

   Annual Meeting of the Association for Computational Linguistics (ACL), 2025

   HF Daily Top-5 Bib PDF Code

   SGG ranked top 5 at Hugging Face Daily Papers on June 3rd, 2025

   @inproceedings{acl2025sgg,
     title = {Taming LLMs by Scaling Learning Rates with Gradient Grouping},
     author = {Li, Siyuan and Tian, Juanxi and Wang, Zedong and Jin, Xin and Liu, Zicheng and Zhang, Wentao and Xu, Dan},
     booktitle = {Annual Meeting of the Association for Computational Linguistics (ACL)},
     year = {2025},
   }

4. 

   Prior-guided Hierarchical Harmonization Network for Efficient Image Dehazing

   Xiongfei Su, Siyuan Li, Yuning Cui, and 7 more authors

   Proceedings of the AAAI Conference on Artificial Intelligence, 2025

   Bib PDF

   @inproceedings{su2025prior,
     title = {Prior-guided Hierarchical Harmonization Network for Efficient Image Dehazing},
     author = {Su, Xiongfei and Li, Siyuan and Cui, Yuning and Cao, Miao and Zhang, Yulun and Chen, Zheng and Wu, Zongliang and Wang, Zedong and Zhang, Yuanlong and Yuan, Xin},
     booktitle = {Proceedings of the AAAI Conference on Artificial Intelligence},
     volume = {39},
     number = {7},
     pages = {7042--7050},
     year = {2025},
   }

2024

1. 

   MogaNet: Multi-order Gated Aggregation Network

   Siyuan Li^*, Zedong Wang^*, Zicheng Liu, and 6 more authors

   International Conference on Learning Representations (ICLR), 2024

   Bib HTML PDF Code

   @inproceedings{iclr2024moganet,
     title = {MogaNet: Multi-order Gated Aggregation Network},
     author = {Li, Siyuan and Wang, Zedong and Liu, Zicheng and Tan, Cheng and Lin, Haitao and Wu, Di and Chen, Zhiyuan and Zheng, Jiangbin and Li, Stan Z.},
     booktitle = {International Conference on Learning Representations (ICLR)},
     year = {2024},
     url = {https://openreview.net/forum?id=XhYWgjqCrV},
   }

2. 

   Unveiling the Backbone-Optimizer Coupling Bias in Visual Representation Learning

   Siyuan Li^*, Juanxi Tian^*, Zedong Wang^*, and 6 more authors

   arXiv preprint arXiv:2410.06373, 2024

   Bib PDF Code

   @article{li2024bocb,
     title = {Unveiling the Backbone-Optimizer Coupling Bias in Visual Representation Learning},
     author = {Li, Siyuan and Tian, Juanxi and Wang, Zedong and Zhang, Luyuan and Liu, Zicheng and Jin, Weiyang and Liu, Yang and Sun, Baigui and Li, Stan Z.},
     journal = {arXiv preprint arXiv:2410.06373},
     year = {2024},
   }

3. 

   Short-Long Convolutions Help Hardware-Efficient Linear Attention to Focus on Long Sequences

   Zicheng Liu, Siyuan Li , Li Wang, and 3 more authors

   International Conference on Machine Learning (ICML), 2024

   Bib PDF Code

   @inproceedings{icml2024chela,
     title = {Short-Long Convolutions Help Hardware-Efficient Linear Attention to Focus on Long Sequences},
     author = {Liu, Zicheng and Li, Siyuan and Wang, Li and Wang, Zedong and Liu, Yunfan and Li, Stan Z.},
     booktitle = {International Conference on Machine Learning (ICML)},
     year = {2024},
     url = {https://openreview.net/forum?id=TRrXkVdhwi},
   }

4. 

   VQDNA: Unleashing the Power of Vector Quantization for Multi-Species Genomic Sequence Modeling

   Siyuan Li^*, Zedong Wang^*, Zicheng Liu, and 5 more authors

   International Conference on Machine Learning (ICML), 2024

   Bib PDF Code

   @inproceedings{icml2024vqdna,
     title = {VQDNA: Unleashing the Power of Vector Quantization for Multi-Species Genomic Sequence Modeling},
     author = {Li, Siyuan and Wang, Zedong and Liu, Zicheng and Wu, Di and Tan, Cheng and Zheng, Jiangbin and Huang, Yufei and Li, Stan Z.},
     booktitle = {International Conference on Machine Learning (ICML)},
     year = {2024},
     url = {https://openreview.net/forum?id=BOunbuapcv},
   }

5. 

   SemiReward: A General Reward Model for Semi-supervised Learning

   Siyuan Li, Weiyang Jin, Zedong Wang, and 4 more authors

   International Conference on Learning Representations (ICLR), 2024

   Bib PDF Code

   @inproceedings{iclr2024semireward,
     title = {SemiReward: A General Reward Model for Semi-supervised Learning},
     author = {Li, Siyuan and Jin, Weiyang and Wang, Zedong and Wu, Fang and Liu, Zicheng and Tan, Cheng and Li, Stan Z.},
     booktitle = {International Conference on Learning Representations (ICLR)},
     year = {2024},
     url = {https://openreview.net/forum?id=dnqPvUjyRI},
   }

6. 

   LongVQ: Long Sequence Modeling with Vector Quantization on Structured Memory

   Zicheng Liu , Li Wang, Siyuan Li, and 3 more authors

   International Joint Conference on Artificial Intelligence (IJCAI), 2024

   Bib PDF Code

   @inproceedings{liu2024longvq,
     title = {LongVQ: Long Sequence Modeling with Vector Quantization on Structured Memory},
     author = {Liu, Zicheng and Wang, Li and Li, Siyuan and Wang, Zedong and Lin, Haitao and Li, Stan Z.},
     booktitle = {International Joint Conference on Artificial Intelligence (IJCAI)},
     year = {2024},
   }

7. 

   Switch EMA: A Free Lunch for Better Flatness and Sharpness

   Siyuan Li, Zicheng Liu, Juanxi Tian, and 9 more authors

   arXiv preprint arXiv:2211.03295, 2024

   Bib PDF

   @article{li2024SEMA,
     title = {Switch EMA: A Free Lunch for Better Flatness and Sharpness},
     author = {Li, Siyuan and Liu, Zicheng and Tian, Juanxi and Wang, Ge and Wang, Zedong and Jin, Weiyang and Wu, Di and Tan, Cheng and Lin, Tao and Liu, Yang and Sun, Baigui and Li, Stan Z.},
     journal = {arXiv preprint arXiv:2211.03295},
     year = {2024},
   }

2023

1. 

   Masked Modeling for Self-supervised Representation Learning on Vision and Beyond

   Siyuan Li, Luyuan Zhang, Zedong Wang, and 8 more authors

   arXiv preprint arXiv:2401.00897, 2023

   Bib PDF Code

   @article{li2023masked,
     title = {Masked Modeling for Self-supervised Representation Learning on Vision and Beyond},
     author = {Li, Siyuan and Zhang, Luyuan and Wang, Zedong and Wu, Di and Wu, Lirong and Liu, Zicheng and Xia, Jun and Tan, Cheng and Liu, Yang and Sun, Baigui and Li, Stan Z.},
     journal = {arXiv preprint arXiv:2401.00897},
     year = {2023},
   }

2. 

   OpenSTL: A Comprehensive Benchmark of Spatio-Temporal Predictive Learning

   Cheng Tan, Siyuan Li, Zhangyang Gao, and 5 more authors

   Advances in Neural Information Processing Systems (NeurIPS), 2023

   Bib PDF Code

   @inproceedings{tan2023openstl,
     title = {OpenSTL: A Comprehensive Benchmark of Spatio-Temporal Predictive Learning},
     author = {Tan, Cheng and Li, Siyuan and Gao, Zhangyang and Guan, Wenfei and Wang, Zedong and Liu, Zicheng and Wu, Lirong and Li, Stan Z.},
     booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
     year = {2023},
   }

2022

1. 

   OpenMixup: Open Mixup Toolbox and Benchmark for Visual Representation Learning

   Siyuan Li^*, Zedong Wang^*, Zicheng Liu, and 2 more authors

   arXiv preprint arXiv:2209.04851, 2022

   Bib PDF Code

   @article{li2022openmixup,
     title = {OpenMixup: Open Mixup Toolbox and Benchmark for Visual Representation Learning},
     author = {Li, Siyuan and Wang, Zedong and Liu, Zicheng and Wu, Di and Li, Stan Z.},
     journal = {arXiv preprint arXiv:2209.04851},
     year = {2022},
   }

2021

1. 

   Boosting Discriminative Visual Representation Learning with Scenario-Agnostic Mixup

   Siyuan Li, Zicheng Liu, Zedong Wang, and 3 more authors

   arXiv preprint arXiv:2111.15454, 2021

   Bib PDF Code

   @article{li2021samix,
     title = {Boosting Discriminative Visual Representation Learning with Scenario-Agnostic Mixup},
     author = {Li, Siyuan and Liu, Zicheng and Wang, Zedong and Wu, Di and Liu, Zihan and Li, Stan Z},
     journal = {arXiv preprint arXiv:2111.15454},
     year = {2021},
   }