CubeComposer: Spatio-Temporal Autoregressive 4K 360° Video Generation from Perspective Video
Lingen Li, Guangzhi Wang*, Xiaoyu Li, Zhaoyang Zhang, Qi Dou, Jinwei Gu, Tianfan Xue, Ying Shan
CVPR, 2026.Paper
Guangzhi Wang
Senior Researcher, Tencent, Shenzhen, China
I am a Senior Researcher at ARC Lab, Tencent PCG, based in Shenzhen. My research focuses on: generative world, agents and their interactions. Prior to this, I obtained my Ph.D. from the National University of Singapore, supervised by Prof. Mohan Kankanhalli. I obtained my bachelor degree from Zhejiang University.
Join Us: We are actively seeking self-motivated research interns to work on Vision-Language-Action (VLA) models and generative environments. We also welcome collaborations of all forms. Reach out via email.
Education
National University of Singapore (NUS)
2020.01 - 2024.05Ph.D. in Data Science
Advisor: Prof. Mohan S. Kankanhalli
Zhejiang University (ZJU)
2015.08 - 2019.06B.E. in Computer Science and Technology
He Zhijun Class | Advisor: Prof. Jianling Sun
News
[Feb 2026]
Two papers (CubeComposer and FlashIn) accepted to CVPR 2026.
[Jan 2026]
Two papers (Gencompositor and Tooncomposer) accepted to ICLR 2026.
[Aug 2025]
One paper (Blobctrl) accepted to SIGGRAPH Asia 2025.
[Feb 2024]
Three papers accepted to CVPR 2024.
[Jul 2023]
Released SEED-Bench, a comprehensive benchmark for Multimodal LLMs.
Selected Publications
Full List* denotes project lead / corresponding author
FlashIn: Fast and Accurate Image Inversion for Real-time Image Editing
Guangzhi Wang
Special thanks to Zhaoyang Zhang, Xiaoyu Li, and Ying Shan for their insightful discussions and support during this project.
CVPR, 2026.Paper
CCRemover: Towards Contextually Consistent Removal of Objects and Visual Effects in Videos
Guangzhi Wang, Xiaoyu Li, Xiaodong Cun, Lingen Li, Ying Shan
Coming Soon, 2026.Paper
EditProp: Consistent Video-to-Video Translation by Editing Propagation
Guangzhi Wang, Xiaoyu Li, Xiaodong Cun, Zhaoyang Zhang, Ying Shan
Coming Soon, 2026.Paper
Tooncomposer: Streamlining cartoon production with generative post-keyframing
Lingen Li, Guangzhi Wang*, Zhaoyang Zhang, Yaowei Li, Xiaoyu Li, Dou Qi, Jinwei Gu, Tianfan Xue, Ying Shan
ICLR, 2026.Paper
Gencompositor: generative video compositing with diffusion transformer
Shuzhou Yang, Xiaoyu Li, Xiaodong Cun, Guangzhi Wang, Lingen Li, Ying Shan, Jian Zhang
ICLR, 2026.Paper
Blobctrl: A unified and flexible framework for element-level image generation and editing
Yaowei Li, Lingen Li, Zhaoyang Zhang, Xiaoyu Li, Guangzhi Wang, Hongxiang Li, Xiaodong Cun, Ying Shan, Yuexian Zou
SIGGRAPH Asia, 2025.Paper
S3Editor: A Sparse Semantic-Disentangled Self-Training Framework for Face Video Editing
Guangzhi Wang, Tianyi Chen, Kamran Ghasedi, Hsiang Tao Wu, Tianyu Ding, Chris Nuesmeyer, Ilya Zharkov, Mohan Kankanhalli, Luming Liang
ArXiv, 2024.Paper
SEED-Bench-2: Benchmarking multimodal large language models
Bohao Li, Yuying Ge, Yixiao Ge, Guangzhi Wang, Rui Wang, Ruimao Zhang, Ying Shan
CVPR, 2024.Paper
SEED-Bench: Benchmarking Multimodal LLMs with Generative Comprehension
Bohao Li*, Rui Wang*, Guangzhi Wang*, Yuying Ge, Yixiao Ge, Ying Shan
ArXiv, 2023.Paper
What makes for good visual tokenizers for large language models?
Guangzhi Wang, Yixiao Ge, Xiaohan Ding, Mohan Kankanhalli, Ying Shan
ArXiv, 2023.Paper
Bilateral Adaptation for Human-Object Interaction Detection with Occlusion-Robustness
Guangzhi Wang, Yangyang Guo, Ziwei Xu, Mohan Kankanhalli
CVPR, 2024.Paper
PELA: Learning parameter-efficient models with low-rank approximation
Yangyang Guo, Guangzhi Wang, Mohan Kankanhalli
CVPR, 2024.Paper
Text to Point Cloud Localization with Relation-Enhanced Transformer
Guangzhi Wang, Hehe Fan, Mohan Kankanhalli
AAAI, 2023.Paper
Distance Matters in Human-Object Interaction Detection
Guangzhi Wang, Yangyang Guo, Yongkang Wong, Mohan Kankanhalli
ACM MM, 2022.Paper
Chairs Can be Stood on: Overcoming Object Bias in Human-Object Interaction Detection
Guangzhi Wang, Yangyang Guo, Yongkang Wong, Mohan Kankanhalli
ECCV, 2022.Paper
Semantic-aware Triplet Loss for Image Classification
Guangzhi Wang, Yangyang Guo, Ziwei Xu, Yongkang Wong, Mohan Kankanhalli
IEEE TMM, 2022.Paper
Multi-source distilling domain adaptation
Guangzhi Wang*, Sicheng Zhao*, Shanghang Zhang, Yang Gu, Yaxian Li, Zhichao Song, Pengfei Xu, Runbo Hu, Hua Chai, Kurt Keutzer
AAAI (Oral), 2020.Paper
Experience
2024 - Present—Senior Researcher, Tencent
Internships
2023.07 - 2023.09—Research Intern, Microsoft Research (Seattle)
2023.01 - 2023.06—Research Intern, Tencent PCG, ARC Lab (Beijing)
2019.10 - 2019.12—Algorithm Intern, Hikvision Research Institute (Hangzhou)
2018.11 - 2019.03—Algorithm Research Intern, DiDi (Beijing)
Academic Service
Conference PC Member/Reviewer: CVPR, ICCV, ECCV, ICLR, NeurIPS, AAAI, ACM MM
Journal Reviewer: IEEE TPAMI, IEEE TMM, IJCV