Publications

2025

1. 

   CVPR’25

   TinyFusion: Diffusion Transformers Learned Shallow

   Gongfan Fang, Kunjun Li, Xinyin Ma, and Xinchao Wang

   Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2025

   Compressing DiTs at 7% Training Costs | 2x Faster Inference

   arXiv Bib Code

   @article{fang2024tinyfusion,
     title = {TinyFusion: Diffusion Transformers Learned Shallow},
     author = {Fang, Gongfan and Li, Kunjun and Ma, Xinyin and Wang, Xinchao},
     journal = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
     year = {2025},
     note = {Compressing DiTs at 7% Training Costs | 2x Faster Inference}
   }

2. 

   CVPR’25

   Collaborative Decoding Makes Visual Auto-Regressive Modeling Efficient

   Zigeng Chen, Xinyin Ma, Gongfan Fang, and Xinchao Wang

   Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2025

   arXiv Bib Code

   @article{chen2024collaborative,
     title = {Collaborative Decoding Makes Visual Auto-Regressive Modeling Efficient},
     author = {Chen, Zigeng and Ma, Xinyin and Fang, Gongfan and Wang, Xinchao},
     journal = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
     year = {2025}
   }

2024

1. 

   NeurIPS’24

   MaskLLM: Learnable Semi-structured Sparsity for Large Language Models

   Gongfan Fang, Hongxu Yin, Saurav Muralidharan, Greg Heinrich, Jeff Pool, Jan Kautz, Pavlo Molchanov, and Xinchao Wang

   Advances in Neural Information Processing Systems, 2024

   NeurIPS Spotlight (2.08%) | Sparse LLMs via End-to-End Training

   arXiv Webpage Bib Code

   @article{fang2024maskllm,
     title = {MaskLLM: Learnable Semi-structured Sparsity for Large Language Models},
     author = {Fang, Gongfan and Yin, Hongxu and Muralidharan, Saurav and Heinrich, Greg and Pool, Jeff and Kautz, Jan and Molchanov, Pavlo and Wang, Xinchao},
     journal = {Advances in Neural Information Processing Systems},
     year = {2024},
     note = {NeurIPS Spotlight (2.08%) | Sparse LLMs via End-to-End Training }
   }

2. 

   NeurIPS’24

   Remix-DiT: Mixing Diffusion Transformers for Multi-Expert Denoising

   Gongfan Fang, Xinyin Ma, and Xinchao Wang

   Advances in Neural Information Processing Systems, 2024

   arXiv Bib Code

   @article{fang2024remixdit,
     title = {Remix-DiT: Mixing Diffusion Transformers for Multi-Expert Denoising},
     author = {Fang, Gongfan and Ma, Xinyin and Wang, Xinchao},
     journal = {Advances in Neural Information Processing Systems},
     year = {2024}
   }

3. 

   ECCV’24

   Isomorphic Pruning for Vision Models

   Gongfan Fang, Xinyin Ma, Michael Bi Mi, and Xinchao Wang

   European Conference on Computer Vision, 2024

   arXiv Bib Code

   @article{fang2024isomorphic,
     title = {Isomorphic Pruning for Vision Models},
     author = {Fang, Gongfan and Ma, Xinyin and Mi, Michael Bi and Wang, Xinchao},
     journal = {European Conference on Computer Vision},
     year = {2024}
   }

4. 

   NeurIPS’24

   AsyncDiff: Parallelizing Diffusion Models by Asynchronous Denoising

   Zigeng Chen, Xinyin Ma, Gongfan Fang, Zhenxiong Tan, and Xinchao Wang

   Advances in Neural Information Processing Systems, 2024

   arXiv Bib Code

   @article{chen2024asyncdiff,
     title = {AsyncDiff: Parallelizing Diffusion Models by Asynchronous Denoising},
     author = {Chen, Zigeng and Ma, Xinyin and Fang, Gongfan and Tan, Zhenxiong and Wang, Xinchao},
     journal = {Advances in Neural Information Processing Systems},
     year = {2024}
   }

5. 

   NeurIPS’24

   Learning-to-Cache: Accelerating Diffusion Transformer via Layer Caching

   Xinyin Ma, Gongfan Fang, Michael Bi Mi, and Xinchao Wang

   Advances in Neural Information Processing Systems, 2024

   arXiv Bib Code

   @article{ma2024learning,
     title = {Learning-to-Cache: Accelerating Diffusion Transformer via Layer Caching},
     author = {Ma, Xinyin and Fang, Gongfan and Mi, Michael Bi and Wang, Xinchao},
     journal = {Advances in Neural Information Processing Systems},
     year = {2024}
   }

6. 

   NeurIPS’24

   SlimSam: 0.1% Data Makes Segment Anything Slim

   Zigeng Chen, Gongfan Fang, Xinyin Ma, and Xinchao Wang

   Advances in Neural Information Processing Systems, 2024

   arXiv Bib Code

   @article{chen20230,
     title = {SlimSam: 0.1\% Data Makes Segment Anything Slim},
     author = {Chen, Zigeng and Fang, Gongfan and Ma, Xinyin and Wang, Xinchao},
     journal = {Advances in Neural Information Processing Systems},
     year = {2024}
   }

7. 

   CVPR’24

   DeepCache: Accelerating Diffusion Models for Free

   Xinyin Ma, Gongfan Fang, and Xinchao Wang

   Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024

   Training-free and almost lossless | 2-7x Speedup on Diffusion Models

   arXiv Bib Code

   @inproceedings{ma2023deepcache,
     title = {DeepCache: Accelerating Diffusion Models for Free},
     author = {Ma, Xinyin and Fang, Gongfan and Wang, Xinchao},
     booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
     year = {2024},
     note = {Training-free and almost lossless | 2-7x Speedup on Diffusion Models}
   }

8. 

   InterSpeech’24

   LiteFocus: Accelerated Diffusion Inference for Long Audio Synthesis

   Zhenxiong Tan, Xinyin Ma, Gongfan Fang, and Xinchao Wang

   Conference of the International Speech Communication Association, 2024

   arXiv Bib Code

   @article{tan2024litefocus,
     title = {LiteFocus: Accelerated Diffusion Inference for Long Audio Synthesis},
     author = {Tan, Zhenxiong and Ma, Xinyin and Fang, Gongfan and Wang, Xinchao},
     journal = {Conference of the International Speech Communication Association},
     year = {2024}
   }

2023

1. 

   CVPR’23

   DepGraph: Towards Any Structural Pruning

   Gongfan Fang, Xinyin Ma, Mingli Song, Michael Bi Mi, and Xinchao Wang

   Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023

   Automated Network Pruning | Top-5 on Github #Model-Compression

   arXiv Bib PDF Code

   @inproceedings{fang2023depgraph,
     title = {DepGraph: Towards Any Structural Pruning},
     author = {Fang, Gongfan and Ma, Xinyin and Song, Mingli and Mi, Michael Bi and Wang, Xinchao},
     booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
     pages = {16091--16101},
     year = {2023},
     note = {Automated Network Pruning | Top-5 on Github #Model-Compression}
   }

2. 

   NeurIPS’23

   LLM-Pruner: On the Structural Pruning of Large Language Models

   Xinyin Ma, Gongfan Fang, and Xinchao Wang

   Advances in Neural Information Processing Systems, 2023

   The First Structured Pruning Method for LLMs | Low-cost Pruning and Training

   arXiv Bib Code

   @inproceedings{ma2023llmpruner,
     title = {LLM-Pruner: On the Structural Pruning of Large Language Models},
     author = {Ma, Xinyin and Fang, Gongfan and Wang, Xinchao},
     booktitle = {Advances in Neural Information Processing Systems},
     year = {2023},
     note = {The First Structured Pruning Method for LLMs | Low-cost Pruning and Training}
   }

3. 

   NeurIPS’23

   Structural Pruning for Diffusion Models

   Gongfan Fang, Xinyin Ma, and Xinchao Wang

   Advances in Neural Information Processing Systems, 2023

   arXiv Bib Code

   @inproceedings{fang2023structural,
     title = {Structural Pruning for Diffusion Models},
     author = {Fang, Gongfan and Ma, Xinyin and Wang, Xinchao},
     booktitle = {Advances in Neural Information Processing Systems},
     year = {2023}
   }

2022

1. 

   AAAI’22

   Up to 100x Faster Data-free Knowledge Distillation

   Gongfan Fang, Kanya Mo, Xinchao Wang, Jie Song, Shitao Bei, Haofei Zhang, and Mingli Song

   Proceedings of the AAAI Conference on Artificial Intelligence, 2022

   arXiv Bib PDF Code

   @inproceedings{fang2022up,
     title = {Up to 100x Faster Data-free Knowledge Distillation},
     author = {Fang, Gongfan and Mo, Kanya and Wang, Xinchao and Song, Jie and Bei, Shitao and Zhang, Haofei and Song, Mingli},
     booktitle = {Proceedings of the AAAI Conference on Artificial Intelligence},
     volume = {36},
     number = {6},
     pages = {6597--6604},
     year = {2022}
   }

2. 

   IJCAI’22

   Prompting to Distill: Boosting Data-Free Knowledge Distillation via Reinforced Prompt

   Xinyin Ma, Xinchao Wang, Gongfan Fang, Yongliang Shen, and Weiming Lu

   Proceedings of International Joint Conference on Artificial Intelligence, 2022

   Bib PDF

   @article{ma2022prompting,
     title = {Prompting to Distill: Boosting Data-Free Knowledge Distillation via Reinforced Prompt},
     author = {Ma, Xinyin and Wang, Xinchao and Fang, Gongfan and Shen, Yongliang and Lu, Weiming},
     journal = {Proceedings of International Joint Conference on Artificial Intelligence},
     year = {2022}
   }

3. 

   TIP’23

   Knowledge Amalgamation for Object Detection with Transformers

   Haofei Zhang, Feng Mao, Mengqi Xue, Gongfan Fang, Zunlei Feng, Jie Song, and Mingli Song

   IEEE Transactions on Image Processing, 2022

   arXiv Bib

   @article{zhang2022knowledge,
     title = {Knowledge Amalgamation for Object Detection with Transformers},
     author = {Zhang, Haofei and Mao, Feng and Xue, Mengqi and Fang, Gongfan and Feng, Zunlei and Song, Jie and Song, Mingli},
     journal = {IEEE Transactions on Image Processing},
     year = {2022}
   }

2021

1. 

   NeurIPS’21

   Mosaicking to Distill: Knowledge Distillation from Out-of-Domain Data

   Gongfan Fang, Yifan Bao, Jie Song, Xinchao Wang, Donglin Xie, Chengchao Shen, and Mingli Song

   Advances in Neural Information Processing Systems, 2021

   arXiv Bib PDF Code

   @article{fang2021mosaicking,
     title = {Mosaicking to Distill: Knowledge Distillation from Out-of-Domain Data},
     author = {Fang, Gongfan and Bao, Yifan and Song, Jie and Wang, Xinchao and Xie, Donglin and Shen, Chengchao and Song, Mingli},
     journal = {Advances in Neural Information Processing Systems},
     volume = {34},
     pages = {11920--11932},
     year = {2021}
   }

2. 

   IJCAI’21

   Contrastive Model Inversion for Data-free Knowledge Distillation

   Gongfan Fang, Jie Song, Xinchao Wang, Chengchao Shen, Xingen Wang, and Mingli Song

   Proceedings of International Joint Conference on Artificial Intelligence, 2021

   arXiv Bib PDF Code

   @article{fang2021contrastive,
     title = {Contrastive Model Inversion for Data-free Knowledge Distillation},
     author = {Fang, Gongfan and Song, Jie and Wang, Xinchao and Shen, Chengchao and Wang, Xingen and Song, Mingli},
     journal = {Proceedings of International Joint Conference on Artificial Intelligence},
     year = {2021}
   }

2020

1. 

   EMNLP’20

   Adversarial Self-Supervised Data-Free Distillation for Text Classification

   Xinyin Ma, Yongliang Shen, Gongfan Fang, Chen Chen, Chenghao Jia, and Weiming Lu

   Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, 2020

   arXiv Bib PDF

   @inproceedings{ma2020adversarial,
     title = {Adversarial Self-Supervised Data-Free Distillation for Text Classification},
     author = {Ma, Xinyin and Shen, Yongliang and Fang, Gongfan and Chen, Chen and Jia, Chenghao and Lu, Weiming},
     booktitle = {Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing},
     pages = {6182--6192},
     year = {2020}
   }

2019

1. 

   Preprint’19

   Data-free Adversarial Distillation

   Gongfan Fang, Jie Song, Chengchao Shen, Xinchao Wang, Da Chen, and Mingli Song

   arXiv preprint arXiv:1912.11006, 2019

   arXiv Bib Code

   @article{fang2019data,
     title = {Data-free Adversarial Distillation},
     author = {Fang, Gongfan and Song, Jie and Shen, Chengchao and Wang, Xinchao and Chen, Da and Song, Mingli},
     journal = {arXiv preprint arXiv:1912.11006},
     year = {2019}
   }