About Me

Hi there! I’m Jianhong Bai. I am currently a Researcher at ByteDance Seed, working on video world models. I received my Ph.D. from Zhejiang University in March 2026, advised by Prof. Haoji Hu. Prior to ByteDance, I was fortunate to work on video generation as a research intern at the Kling Team, Kuaishou (2024–2026) with Menghan Xia and Xintao Wang, and at Microsoft Research Asia (2023–2024) with Tianyu He and Xu Tan. Feel free to reach out to me at cpurgicn@gmail.com.

I am looking for self-motivated research interns to work on video world models and video generation. Please drop me an email if interested.

Recent News

• 2026/03: Joined ByteDance Seed as a Researcher, working on video world models.
• 2026/03: Received my Ph.D. from Zhejiang University.
• 2026/02: Concluded my research internship at the Kling Team, Kuaishou.
• 2025/12: Released SemanticGen, a video generation framework that operates in the semantic space, on arXiv.
• 2025/10: Our work ReCamMaster is selected as ICCV'25 Best Paper Finalist.
• 2025/08: 2 paper accepted by SIGGRAPH ASIA'25.
• 2025/07: 1 paper accepted by ICCV'25.
• 2025/01: 1 paper accepted by ICLR'25.
• 2024/12: 1 paper accepted by AAAI'25.
• 2024/05: Start a research internship at Kling Team, advised by Prof. Menghan Xia and Prof. Xintao Wang.
• 2023/12: 1 paper accepted by AAAI'24.
• 2023/09: 2 paper accepted by NeurIPS'23.
• 2023/08: Start a research internship at Microsoft Research Asia (MSRA), advised by Prof. Tianyu He and Prof. Xu Tan.
• 2023/01: 1 paper accepted by ICLR'23.
• 2021/09: Join Prof. Hu's lab at Zhejiang University.

Selected Publications

	Jianhong Bai, Xiaoshi Wu, Xintao Wang et al. "SemanticGen: Video Generation in Semantic Space". Preprint 2025. Arxiv Project page
	Jianhong Bai, Menghan Xia, Xintao Wang et al. "ReCamMaster: Camera-Controlled Generative Rendering from A Single Video". In ICCV 2025, Best Paper Finalist. Arxiv Project page
	Jianhong Bai, Menghan Xia, Xintao Wang et al. "SynCamMaster: Synchronizing Multi-Camera Video Generation from Diverse Viewpoints". In ICLR 2025. Arxiv Project page
	Yawen Luo, Jianhong Bai, Xiaoyu Shi et al. "CamCloneMaster: Enabling Reference-based Camera Control for Video Generation". In SIGGRAPH ASIA 2025. Arxiv Project page
	Jiwen Yu, Jianhong Bai, Yiran Qin et al. "Context as Memory: Scene-Consistent Interactive Long Video Generation with Memory Retrieval". In SIGGRAPH ASIA 2025. Arxiv Project page
	Jianhong Bai, Tianyu He, Yuchi Wang et al. "UniEdit: A Unified Tuning-Free Framework for Video Motion and Appearance Editing". In ACM MM 2025. Arxiv Code Project page
	Xiao Fu, Xintao Wang, Xian Liu, Jianhong Bai et al. "Learning Video Generation for Robotic Manipulation with Collaborative Trajectory Control". Preprint 2025. Arxiv Code Project page
	Yuchi Wang, Junliang Guo, Jianhong Bai et al. "InstructAvatar: Text-Guided Emotion and Motion Control for Avatar Generation". In AAAI 2024. Arxiv Project page
	Jianhong Bai, Zuozhu Liu, Hualiang Wang et al. "Towards Distribution-Agnostic Generalized Category Discovery". In NeurIPS 2023. Arxiv Code
	Jianhong Bai, Zuozhu Liu, Hualiang Wang et al. "On The Effectiveness of Out-of-Distribution Data in Self-Supervised Long-Tail Learning". In ICLR 2023. Arxiv Code
	Jianhong Bai, Yuchen Yang, Huanpeng Chu et al. "Robustness-Guided Image Synthesis for Data-Free Quantization". In AAAI 2024. Arxiv
	Lianrui Mu*, Jianhong Bai*, Xiaoxuan He et al. "TeG-DG: Textually Guided Domain Generalization for Face Anti-Spoofing". Preprint 2023. Arxiv

Experience

	Researcher, ByteDance Seed Mar. 2026 - Present Advisor: Haoqi Fan Interests: Video World Models
	Research Intern, Kling, Kuaishou May. 2024 - Feb. 2026 Advisor: Menghan Xia, Xintao Wang Interests: Video Generation
	Research Intern, Machine Learning Group, Microsoft Research Asia Aug. 2023 - May. 2024 Advisor: Tianyu He, Xu Tan Interests: Video Generation
	Ph.D., Zhejiang University Sep. 2021 - Mar. 2026 Advisor: Haoji Hu Interests: Computer Vision