Yunshan Zhong (钟云山)

I am currently an Associate Researcher (副研究员) at the School of Computer Science and Technology, Hainan University (海南大学计算机科学与技术学院). I received my Ph.D. from the MAC Lab, Xiamen University, advised by Prof. Rongrong Ji. Before that, I obtained my M.S. from Peking University and B.S. from Beijing Institute of Technology.

I am actively recruiting motivated undergraduate students at Hainan University who are passionate about AI research to join my group. I also welcome prospective M.S. students to apply.

I provide every member of my group with systematic and rigorous research training — from literature review and experimental design to paper writing — with hands-on mentorship at every stage to build a solid academic foundation. The group has ample computing resources and covers AI API costs (e.g., OpenAI, Google Gemini, Claude), so you can focus on research without logistical concerns.

For outstanding students, I will proactively recommend you to top tech companies (e.g., ByteDance, Tencent, Alibaba, Huawei) for core algorithm team internships or full-time positions, as well as to 985 universities for M.S. or Ph.D. programs. I will assist in connecting you with suitable advisors and writing recommendation letters. Several students I have previously recommended have gone on to achieve great success at top companies and universities.

If you are passionate about AI and ready to dive deep, join us — let’s do impactful work together!

I have published around 20 papers in top-tier conferences and journals (CCF-A/B), with 10+ as first or corresponding author. My research focuses on making state-of-the-art AI models efficient and deployable, spanning efficient LLMs, efficient visual generative models, and efficient visual foundation models, with additional interests in bioinformatics computing.

🔬 Research Interests

My research primarily focuses on efficient LLMs, efficient visual generative models, and efficient visual foundation models. Specific directions include:

• LLM Compression & Acceleration
• Video Generation & Understanding Model Acceleration
• VLA (Vision-Language-Action) Model Acceleration
• Efficient Model Safety

I am also interested in bioinformatics computing, including:

• Biological Language Models
• Biological Agents

📢 News

• 2026.03 — One paper accepted to CVPR 2026 (Findings) (Data-Free Quantization for CLIP)
• 2026.01 — One paper accepted to ICLR 2026 (Test-Time Error Correction for Diffusion Models)
• 2025.07 — Two papers accepted to ICCV 2025 (PTQ for SAM, Data-Free Quantization for ViTs)
• 2025 — Two papers published in TPAMI (I&S-ViT, Accurate PTQ of ViTs)
• 2024 — One paper accepted to ICML 2024 (Spotlight, 3.5%) — ERQ for PTQ of ViTs
• 2024 — One paper accepted to AAAI 2024 (Image Demoireing from Unpaired Data)

📝 Selected Publications

A full list is available on the Publications page and Google Scholar. † denotes corresponding author.

2026

• D4C: Data-Free Quantization for Contrastive Language-Image Pre-training Models, CVPR 2026 (Findings), CCF-A. Authors: Wenlun Zhang, Yunshan Zhong†, Zihao Ding, Xinyu Li, Kentaro Yoshioka.

• Test-Time Iterative Error Correction for Efficient Diffusion Models, ICLR 2026, CCF-A. Authors: Yunshan Zhong, Weiqi Yan, Yuxin Zhang.

2025

• I&S-ViT: An Inclusive & Stable Method for Pushing the Limit of Post-Training ViTs Quantization, IEEE TPAMI, CCF-A. Authors: Yunshan Zhong, Jiawei Hu, Mingbao Lin, Mengzhao Chen, Rongrong Ji†.

• Towards Accurate Post-Training Quantization of Vision Transformers via Error Reduction, IEEE TPAMI, CCF-A. Authors: Yunshan Zhong, You Huang, Jiawei Hu, Yuxin Zhang, Rongrong Ji†.

• Distribution-Flexible Subset Quantization for Post-Quantizing Super-Resolution Networks, SCIENCE CHINA Information Sciences (SCIS), CCF-A. Authors: Yunshan Zhong, Mingbao Lin, Jingjing Xie, Yuxin Zhang, Fei Chao, Rongrong Ji†.

• AHCPTQ: Accurate and Hardware-Compatible Post-Training Quantization for SAM, ICCV 2025, CCF-A. Authors: Wenlun Zhang, Yunshan Zhong†, Shimpei Ando, Kentaro Yoshioka.

• Semantic Alignment and Reinforcement for Data-Free Quantization of Vision Transformers, ICCV 2025, CCF-A. Authors: Yunshan Zhong, Yuyao Zhou, Yuxin Zhang, Wanchen Sui, Shen Li, Yong Li, Fei Chao, Rongrong Ji†.

• MultiQuant: A Novel Multi-Branch Topology Method for Arbitrary Bit-width Network Quantization, Pattern Recognition (PR), CCF-B. Authors: Yunshan Zhong, Yuyao Zhou, Fei Chao, Rongrong Ji†.

2024

• ERQ: Error Reduction for Post-Training Quantization of Vision Transformers, ICML 2024 (Spotlight, 3.5%), CCF-A. Authors: Yunshan Zhong, Jiawei Hu, You Huang, Yuxin Zhang, Rongrong Ji†.

• Learning Image Demoireing from Unpaired Real Data, AAAI 2024, CCF-A. Authors: Yunshan Zhong, Yuyao Zhou, Yuxin Zhang, Fei Chao, Rongrong Ji†.

2022

• Dynamic Dual Trainable Bounds for Ultra-low Precision Super-Resolution Networks, ECCV 2022, CCF-B. Authors: Yunshan Zhong, Mingbao Lin, Xunchao Li, Ke Li, Yunhang Shen, Fei Chao, Yongjian Wu, Rongrong Ji†.

• Fine-grained Data Distribution Alignment for Post-Training Quantization, ECCV 2022, CCF-B. Authors: Yunshan Zhong, Mingbao Lin, Mengzhao Chen, Ke Li, Yunhang Shen, Fei Chao, Yongjian Wu, Rongrong Ji†.

🎓 Education

💼 Experience

• 2025.06 – Present — Associate Researcher (副研究员), School of Computer Science and Technology, Hainan University
• 2021.09 – 2025.06 — Ph.D. candidate, MAC Lab, Xiamen University
• 2022.05 – 2023.05 — Research Intern, Pengcheng Lab, Shenzhen
• 2020.11 – 2021.09 — Research Intern, MAC Lab, Xiamen University
• 2020.07 – 2020.11 — R&D Engineer, Baidu, Beijing
• 2019.11 – 2020.02 — Research Intern, Kuaishou Technology, Beijing
• 2018.09 – 2019.09 — Research Intern, Megvii Technology (Face++), Beijing

🔍 Professional Service

Reviewer: ICML, ICLR, NeurIPS, CVPR, ICCV, ECCV, IEEE TNNLS, IEEE TCSVT