Mushui Liu 刘木水

📝 Selected Publications

Full publication list can be found on Google Scholar.
^# co-first author | ^* corresponding author.

1. Image Generation

1. TFCustom: Customized Image Generation with Time-Aware Frequency Feature Guidance [Paper]
   Mushui Liu^#, Dong She^#, Jingxuan Pang, Qihan Huang, Jiacheng Ying, Wanggui He, Yuanlei Hou, Siming Fu^*
   IEEE Conference on Computer Vision and Pattern Recognition (CVPR, Highlight ⭐⭐⭐) , 2025.
2. LLM4GEN: Leveraging Semantic Representation of LLMs for Text-to-Image Generation [Paper]
   Mushui Liu^#, Yuhang Ma^#, Zhen Yang, Jun Dan, Yunlong Yu^*, Zeng Zhao^*, Bai Liu, Changjie Fan, Zhipeng Hu
   The 39th Annual AAAI Conference on Artificial Intelligence (AAAI). 2025.
3. CoAR: Concept Injection into Autoregressive Models for Personalized Text-to-Image Generation [Paper]
   Fangtai Wu^#, Mushui Liu^#, Weijie He, Wanggui He, Hao Jiang, Zhao Wang, Yunlong Yu^*
   In Submission.
4. MOSAIC: Multi-Subject Personalized Generation via Correspondence-Aware Alignment and Disentanglement [Paper]
   Dong She^#, Siming Fu^#, Mushui Liu^#, Qiaoqiao Jin, Hualiang Wang, Mu Liu, Jidong Jiang
   In Submission.
5. RestorerID: Towards Tuning-Free Face Restoration with ID Preservation [Paper]
   Jiacheng Ying^#, Mushui Liu^#, Zhaoyang Wu, Rui Zhang, Zhelun Yu, Siming Fu, Shiyu Cao, Chen Wu, Yunlong Yu, Hailin Shen^*
   In Submission.
6. RectifiedHR: Enable Efficient High-Resolution Image Generation via Energy Rectification [Paper]
   Zhen Yang, Guibao Shen, Liang Hou, Mushui Liu, Luozhou Wang, Xin Tao, Pengfei Wan, Di Zhang, Ying-Cong Chen^*
   In Submission.

2. Unified Models

1. FUSE: Fine-Grained and Semantic-Aware Learning for Unified Image Understanding and Generation
   Peng Zhang^#, Wanggui He^#, Mushui Liu^#, Wenyi Xiao, Siyu Zou, Yuan Li, Xingjian Wang, Guanghao Zhang, Yanpeng Liu, Weilong Dai, Jinlong Liu, Shuyi Ying, Ruikai Zhou, Yunlong Yu, Yubo Tao, Hai Lin^*, Hao Jiang^*
   The 40th Annual AAAI Conference on Artificial Intelligence (AAAI). 2026.
2. Mars: Mixture of Auto-Regressive Models for Fine-grained Text-to-Image Synthesis [Paper]
   Wanggui He^#, Siming Fu^#, Mushui Liu^#, Xierui Wang, Wenyi Xiao, Fangxun Shu, Yi Wang, Lei Zhang, Zhelun Yu, Haoyuan Li, Ziwei Huang, LeiLei Gan^*, Hao Jiang^*
   The 39th Annual AAAI Conference on Artificial Intelligence (AAAI). 2025.
3. Mint: Multi-Modal Chain of Thought in Unified Generative Models for Enhanced Image Generation [Paper]
   Yi Wang^#, Mushui Liu^#, Wanggui He^#, Lei Zhang, Ziwei Huang, Guanghao Zhang, Fangxun Shu, Yubo Tao, ...
   In Submission.

3. Video Understanding & Generation

1. CMMCoT: Enhancing Complex Multi-Image Comprehension via Multi-Modal Chain-of-Thought and Memory Augmentation [Paper]
   Guozhen Zhang^#, Tao Zhong^#, Yan Xia^#, Mushui Liu^#, Zhelun Yu, Haoyuan Li, Wanggui He, Fangxun Shu, ...
   The 40th Annual AAAI Conference on Artificial Intelligence (AAAI). 2026.
2. Frame Order Matters: A Temporal Sequence-Aware Model for Few-Shot Action Recognition [Paper]
   Bozheng Li^#, Mushui Liu^#, Gaoang Wang, Yunlong Yu^*
   The 39th Annual AAAI Conference on Artificial Intelligence (AAAI). 2025.
3. OmniCLIP: Adapting CLIP for Video Recognition with Spatial-Temporal Omni-Scale Feature Learning [Paper]
   Mushui Liu, Bozheng Li, Yunlong Yu^*
   European Conference on Artificial Intelligence (ECAI). 2024.
4. DyST-XL: Dynamic Layout Planning and Content Control for Compositional Text-to-Video Generation [Paper]
   Mushui Liu^#, Weijie He^#, Yunlong Yu^*, Zhao Wang, Chao Wu
   In Submission.
5. CustomVideoX: 3D Reference Attention Driven Dynamic Adaptation for Zero-Shot Customized Video Diffusion Transformers [Paper]
   Dong She^#, Mushui Liu^#, Jingxuan Pang, Jin Wang, Zhen Yang, Wanggui He, Guanghao Zhang, Yi Wang, ...
   In Submission.

4. Representation Learning

1. Envisioning Class Entity Reasoning by Large Language Models for Few-shot Learning [Paper]
   Mushui Liu, Fangtai Wu, Bozheng Li, Ziqian Lu, Yunlong Yu^*, Xi Li
   The 39th Annual AAAI Conference on Artificial Intelligence (AAAI). 2025.
2. Improving Zero-Shot Generalization for CLIP with Variational Adapter [Paper]
   Ziqian Lu, Fangtai Shen, Mushui Liu, Yunlong Yu^*, Zhao Wang, Xi Li, Jungong Han
   European Conference on Computer Vision (ECCV). 2024.
3. Variational Adapter: Improving CLIP in Data-Imbalanced Scenarios [Paper]
   Ziqian Lu, Mushui Liu, Yunlong Yu^*, Zhao Wang, Xi Li, Jungong Han
   IEEE Transactions on Circuits and Systems for Video Technology (IEEE TCSVT). 2025.
4. Synth-CLIP: Synthetic Data Make CLIP Generalize Better in Data-Limited Scenarios [Paper]
   Mushui Liu, Weijie He, Ziqian Lu, Jun Dan, Yunlong Yu^*, Yingming Li, Xi Li, Jungong Han
   Neural Networks (Neural Networks). 2025.
5. Fully Fine-Tuned CLIP Models are Efficient Few-Shot Learners [Paper]
   Mushui Liu, Bozheng Li, Jun Dan, Ziqian Lu, Zhao Wang, Yunlong Yu^*
   Knowledge-Based Systems (KBS). 2025.
6. Tolerant Self-Distillation for Image Classification [Paper]
   Mushui Liu, Yunlong Yu^*, Zhong Ji, Jungong Han, Zhongfei Zhang
   Neural Networks (Neural Networks). 2024.
7. Hybrid mask generation for infrared small target detection with single-point supervision [Paper]
   Weijie He, Mushui Liu, Yunlong Yu^*
   Neurocomputing (Neurocomputing). 2025.
8. Lightweight MIMO-WNet for single image deblurring [Paper]
   Mushui Liu, Yunlong Yu^*, Yingming Li, Zhong Ji, Wen Chen, Yang Peng
   Neurocomputing (Neurocomputing), Vol. 516, pp. 106-114. 2023.
9. CM-UNet: Hybrid CNN-Mamba UNet for Remote Sensing Image Semantic Segmentation [Paper]
   Mushui Liu, Jun Dan, Ziqian Lu, Yunlong Yu^*, Yuhang Li, Xi Li
   In Submission.