Junkun Yuan   袁俊坤

Research Scientist, ByteDance (US)

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

yuanjk0921@outlook.com

Junkun Yuan

Biography    Publications    Professional Service

Biography

I am a Research Scientist at ByteDance (US), working on visual generative foundation models, such as Seedance 2.0, and their applications and products.

During 2023–2025, I worked as a Research Scientist in the Hunyuan Multimodal Generation Team at Tencent with Wei Liu, Zhao Zhong, and Liefeng Bo, where my research focused on multimodal generative foundation models and downstream generation tasks. During 2022–2023, I was a research intern in the Computer Vision Group at Baidu with Xinyu Zhang and Jingdong Wang, where my research focused 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 Zhejiang University of Technology (2015–2019), supervised by Professor Qi Xuan.

I have been fortunate to work closely with friends including Defang Chen and Yue Ma; their insights have profoundly shaped my approach to research.

Selected Publications

Google Scholar Profile    Semantic Scholar Profile      Full Publication List →

(co-)first author✳    corresponding author✉

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 Generation Team at Tencent (as a group member)

arXiv, 2024

Dec 03, 2024   |   HunyuanVideo   |   code

It introduces an open-source diffusion model for video generation, which has received over 1000 citations and over 11,000 GitHub stars (as of May 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

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

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

Professional Service

Last updated on May 14, 2026 at 21:07 (UTC-7)