Junkun Yuan   袁俊坤

Senior Research Scientist, ByteDance (US)

Work and live in San Jose (US)  &  Shenzhen (China)

yuanjk0921@outlook.com

Junkun Yuan

Biography    Selected Publications    Professional Service

Biography

I am a Senior Research Scientist at ByteDance (US), working with Chongyang Ma. My work focuses on visual generative foundation models, including Seedance 2.0/2.5, and translating them into applications and products. My research interests span generative AI, computer vision, and deep learning.

During 2023–2025, I worked as a Research Scientist at the Hunyuan Multimodal Foundation Model Team at Tencent, working with Wei Liu and Zhao Zhong on large-scale visual generative foundation models, such as HunyuanVideo, and a range of downstream generation tasks. During 2022–2023, I was a research intern in the Computer Vision Group at Baidu, working with Xinyu Zhang and Jingdong Wang on visual self-supervised pre-training.

I received my Ph.D. degree in Computer Science from Zhejiang University (2019–2024), co-supervised by Professors Kun Kuang, Lanfen Lin, and Fei Wu. I received my B.E. degree in Automation from Jianxing Honors College, Zhejiang University of Technology (2015–2019), supervised by Professor Qi Xuan.

Selected Publications

Full Publication List →    Google Scholar Profile    Semantic Scholar Profile

(co-)first author✳    corresponding author/project lead✉

Follow-Your-Preference: Towards Preference-Aligned Image Inpainting

Yutao Shen✳, Junkun Yuan✳✉, Toru Aonishi, Hideki Nakayama, et al.

International Conference on Learning Representations (ICLR), 2026

Sep 27, 2025      Follow-Your-Preference      code

HunyuanVideo: A Systematic Framework For Large Video Generative Models

Hunyuan Multimodal Foundation Model Team at Tencent (as a team member)

Technical Report, 2024

Dec 03, 2024      HunyuanVideo      code

It introduces an open-source diffusion model for video generation, which has over 1,500 citations and over 12,000 GitHub stars (as of Aug 2026).

Follow-Your-Canvas: Higher-Resolution Video Outpainting with Extensive Content Generation

Qihua Chen✳, Yue Ma✳, Hongfa Wang✳, Junkun Yuan✳✉, et al.

AAAI Conference on Artificial Intelligence (AAAI), 2025

Sep 02, 2024      Follow-Your-Canvas      code

HAP: Structure-Aware Masked Image Modeling for Human-Centric Perception

Junkun Yuan✳, Xinyu Zhang✳✉, Hao Zhou, Jian Wang, et al.

Advances in Neural Information Processing Systems (NeurIPS), 2023

Oct 31, 2023      HAP      code

Label-Efficient Domain Generalization via Collaborative Exploration and Generalization

Junkun Yuan✳, Xu Ma✳, Defang Chen, Kun Kuang✉, et al.

International Conference on Multimedia (MM), 2022

Aug 07, 2022      CEG      code

Collaborative Semantic Aggregation and Calibration for Federated Domain Generalization

Junkun Yuan✳, Xu Ma✳, Defang Chen, Fei Wu, et al.

IEEE Transactions on Knowledge and Data Engineering (TKDE), 2023

Oct 13, 2021      CSAC      code

Domain-Specific Bias Filtering for Single Labeled Domain Generalization

Junkun Yuan✳, Xu Ma✳, Defang Chen, Kun Kuang✉, et al.

International Journal of Computer Vision (IJCV), 2022

Oct 02, 2021      DSBF      code

Professional Service

Last updated on September 17, 2026  ·  Design inspired by Kaiming He's homepage.