I am a Ph.D. student in Computer Science and Technology at Shanghai Jiao Tong University, advised by Prof. Xiaohong Liu and jointly trained at Shanghai Innovation Institute. Previously, I received my B.Eng. in Software Engineering from Central South University. My research interests include AIGC, video generation, video editing, and multimodal generation.

📖 Education

• 2026.09 - Present: Jointly Trained Ph.D. Student, Shanghai Innovation Institute.
• 2026.09 - Present: Ph.D. Student, Computer Science and Technology, Shanghai Jiao Tong University.
• 2022.09 - 2026.06: B.Eng., Software Engineering, Central South University.

🔥 News

• [06/2026] One paper has been accepted by European Conference on Computer Vision (ECCV) 2026.
• [06/2026] One paper has been accepted by IEEE Transactions on Circuits and Systems for Video Technology (TCSVT).
• [05/2026] Two papers have been submitted to NeurIPS 2026. Wish us luck!
• [05/2026] One paper has been accepted by ICML 2026.
• [03/2026] Excited to release A²Edit, a reference-image-guided framework for image restoration and editing with powerful generative capabilities. It has been submitted to ECCV 2026.
• [02/2026] Excited to release LayerT2V, a unified multi-layer video generation framework. We build the first large-scale multi-layer video dataset. It has been submitted to ICML 2026.
• [02/2026] One paper has been accepted by CVPR 2026.
• [12/2025] Excited to release FlowDirector, a training- and inversion-free video editing framework that demonstrates strong performance. It has been submitted to CVPR 2026.
• [05/2025] One paper has been submitted to TCSVT.

📝 Publications

* equal contribution, † corresponding author

From Redundancy to Efficiency: Adaptive Video Tokenization via Appearance-Motion Decoupling and Dynamic Token Allocation

Guangzhao Li^*, Qingyan Wei^*, Yan Tai, Chaoyang Zhang, Linfeng Zhang, Zuxuan Wu, Guangtao Zhai, Xiaohong Liu^†

Submitted to Conference on Neural Information Processing Systems (NeurIPS) 2026

RewardFlow: Learning to Align Flow Models with Reward Distributions

Qingyan Wei^*, Guangzhao Li^*, Xiaohong Liu, Xiaobing Tu, Jinkui Ren, Xiantao Zhang, Peiyuan Chen, Linfeng Zhang^†

Submitted to Conference on Neural Information Processing Systems (NeurIPS) 2026

A²-Edit: Precise Reference-Guided Image Editing of Arbitrary Objects and Ambiguous Masks

Huayu Zheng, Guangzhao Li, Baixuan Zhao, Siqi Luo, Hantao Jiang, Guangtao Zhai, Xiaohong Liu^†

European Conference on Computer Vision (ECCV) 2026

[paper] [code] [project page]

LayerT2V: A Unified Multi-Layer Video Generation Framework

Guangzhao Li^*, Kangrui Cen^*, Baixuan Zhao, Yi Xin, Siqi Luo, Guangtao Zhai, Lei Zhang, Xiaohong Liu^†

International Conference on Machine Learning (ICML) 2026

[paper] [code] [project page]

FlowDirector: Training-Free Flow Steering for Precise Text-to-Video Editing

Guangzhao Li, Yanming Yang, Chenxi Song, Xiaohong Liu, Chi Zhang^†

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2026

[paper] [code] [project page]

SceneVLP: Augmenting Vision-Language Pre-Training Models for Video Action Recognition with Video Scene Graphs

Liangzhen Liu^*, Guangzhao Li^*, Qi Xie, Zhaowen Wang, Li Kuang^†, Yingjie Xia, Jianwei Yin

IEEE Transactions on Circuits and Systems for Video Technology (TCSVT)