Yiming Zhang

Yiming Zhang 张倚铭

I am currently a PhD. candidate at USTC (中国科学技术大学), advised by Yanhong Zeng and Bin Liu. Before that, I received my bachelor's degree of B.Eng. from Dalian University of Technology in 2024.

My research interests focus on Generative AI, with research experience in Pretraining, Post-training and Agentic Training of foundation models.

Email / Google Scholar / Twitter / Github

News

[2026.08] Intern-S2-Preview has been released, with pretraining supported by our Visual Pretraining.
[2026.07] Release the preprint of Scalable Visual Pretraining.
[2026.02] Our paper Exploring Visual Pretraining is accepted by CVPR 2026.
[2026.01] Our paper InternGeometry is accepted by ICLR 2026.
[2025.10] Our paper FoleyCrafter is accepted by IJCV.
[2024.07] Release the paper and code of FoleyCrafter.
[2024.06] Graduated from Dalian University of Technology.
[2024.03] Our paper PIA is accepted by CVPR 2024.

Research

	Scalable Visual Pretraining for Language Intelligence Yiming Zhang, Zhonghan Zhao, Wenwei Zhang, Haiteng Zhao, Tianyang Lin, Huanze Tang, Yunhua Zhou, Demin Song, Kuikun Liu, Haochen Ye, Haian Huang, Yuzhe Gu, Haijun Lv, Qipeng Guo, Bin Liu, Gaoang Wang, Kai Chen arXiv preprint, 2026 arXiv A systematic study showing that visual pretraining directly on rendered documents consistently outperforms text-only pretraining on the same corpora, offering an efficient pathway to scalable language intelligence.
	FoleyCrafter: Bring Silent Videos to Life with Lifelike and Synchronized Sounds Yiming Zhang, Yicheng Gu, Yanhong Zeng, Zhening Xing, Yuancheng Wang, Zhizheng Wu, Kai Chen IJCV, 2026 project page / video / arXiv / demo / code FoleyCrafter is a video-to-audio generation framework which can produce realistic sound effects semantically relevant and synchronized with videos.
	PIA: Your Personalized Image Animator via Plug-and-Play Modules in Text-to-Image Models Yiming Zhang, Zhening Xing, Yanhong Zeng, Youqing Fang, Kai Chen CVPR, 2024 project page / video / arXiv / demo / code PIA can animate any images from personalized models by text while preserving high-fidelity details and unique styles.
	Exploring Visual Pretraining for Learning Language Intelligence Zhonghan Zhao, Yiming Zhang, Wenwei Zhang, Haiteng Zhao, Xingguang Wei, Zhangwei Gao, Kuikun Liu, Yuzhe Gu, Size Wu, Haian Huang, Jianfei Gao, Haijun Lv, Demin Song, Yunhua Zhou, Qipeng Guo, Gaoang Wang, Kai Chen CVPR, 2026 paper A first attempt (MAPLE) showing that LLMs can be pretrained on visual images of documents to reach parity with text-only models, breaking the text-scaling bottleneck.
	Achieving Olympiad-Level Geometry Large Language Model Agent via Complexity Boosting Reinforcement Learning Haiteng Zhao, Junhao Shen, Yiming Zhang, Songyang Gao, Kuikun Liu, Tianyou Ma, Fan Zheng, Dahua Lin, Wenwei Zhang, Kai Chen ICLR*, 2026 arXiv InternGeometry is an LLM agent achieving medalist-level performance on olympiad geometry via complexity-boosting reinforcement learning, using only 0.04% of the data used by AlphaGeometry 2.

Project

Intern-S2-Preview: Scientific Agentic Foundation Model
Lei Bai, Jiaqi Cao, Chiyu Chen, ..., Yiming Zhang, ..., Kai Chen, et al.
arXiv preprint, 2026
arXiv

Intern-S2-Preview is a series of scientific agentic foundation models supporting multimodal scientific understanding, reasoning, generation, and long-horizon agentic tasks.

Working Experience

Shanghai AI Laboratory

Researcher Intern, 2023.07 ~ Present

Awards

Invited Talk: Personalized Image Animator (OpenMMLab on Bilibili Live 2024)
Conference Reviewer: NeurIPS, CVPR, ICCV, ECCV.
Outstanding Undergraduate in 2024.
National Scholarship in 2022 (Top 1% in DUT).
Merit Students.
First Prize Excellence Scholarship.