Scholar
Junkun Yuan
Google Scholar ID: j3iFVPsAAAAJ
Research Scientist, Tencent
Citations & Impact
Citations
1,192
H-index
16
i10-index
21
Publications
20
Co-authors
7
list available
Contact
Publications
Resume (English only)
Academic Achievements
- Published several papers, including AsynDM (arXiv 2025), Follow-Your-Preference (arXiv 2025), Follow-Your-Emoji-Faster (IJCV 2025), and Hunyuan-Game (arXiv 2025). Notably, HunyuanVideo is a systematic framework for large video generative models, featuring an open-source diffusion model with 13B parameters, which has over 500 citations and over 11,000 GitHub stars as of November 2025.
Research Experience
- Sep 2023 - Nov 2025, interned and then worked in the Hunyuan Multimodal Generation Foundation Model Team at Tencent, collaborating with Wei Liu, Liefeng Bo, and Zhao Zhong; Jul 2022 - Aug 2023, interned in the Computer Vision Group at Baidu, collaborating with Xinyu Zhang and Jingdong Wang.
Education
- 2019-2024, Ph.D. in Computer Science from Zhejiang University, supervised by Professors Kun Kuang, Lanfen Lin, and Fei Wu; 2015-2019, B.E. in Automation from Zhejiang University of Technology, supervised by Professors Qi Xuan and Li Yu.
Background
- Research interests include multimodal generative foundation models and visual self-supervised pre-training. Currently working in the Hunyuan Multimodal Generation Foundation Model Team at Tencent, focusing on multimodal generative foundation models.
Miscellany
- Worked closely with friends such as Defang Chen and Yue Ma, whose insights have profoundly shaped his approach to research.
