arXiv 2026First author
AniMatrix: An Anime Video Generation Model that Thinks in Art, Not Physics
X Wen, Tencent HY Team
- TPAMI / ICMLJournal and conference
- SIGGRAPH Asia / ACM MMGraphics and multimedia
- GDC / GTCInvited talks
- SkyReels-V12.7K stars
About
I am working toward real-time interactive virtual worlds: environments generated as you move through them, that change because of what you did, rather than finished before you arrive.
That problem sits across generative AI, diffusion models, computer graphics, and 3D vision. I have shipped work in all four, and the hard parts tend to fall in the seams between them. Right now that means post-training video generation models at Tencent Hunyuan: generation quality, a unified architecture handling text, image, multi-reference, and editing in one model, and the inference efficiency that real-time interaction ultimately depends on.
Before Tencent I built SkyReels AI, an AI short-drama platform, from scratch as its business lead and CTO, led game AI at ByteDance, and worked on game agents and content generation at NetEase Fuxi AI Lab. Over the same years I finished a PhD in artificial intelligence at Zhejiang University.
News
- 2026.03Received the Business Application Pioneer Award at Tencent for leading the algorithm work on AniMatrix.Award
- 2026.03HY-WU open-sourced at Tencent Hunyuan, a framework that generates instance-conditioned LoRA adapters on the fly for text-guided image editing.Open sourcearXiv
- 2025.12Completed my PhD in artificial intelligence at Zhejiang University.
- 2025.08VolGen accepted to TPAMI.TPAMI
- 2025.07Joined the Tencent Hunyuan foundation model department, working on post-training for video generation.
- 2025.02Open-sourced SkyReels-V1, a human-centric video foundation model, releasing both text-to-video and image-to-video weights along with the inference framework.Open source
- 2025.02Released SkyReels-A1, expressive portrait animation on video diffusion transformers, with code.arXivOpen source
- 2024.12Released Video Diffusion Transformers are In-Context Learners.arXiv
- 2024.09Talk at the Apsara Conference on AI workflows for producing drama online, from script generation to storyboarding.Apsara
- 2024.08Released SkyScript-100M, a script and shooting-script dataset for short drama.arXiv
- 2023.07Joined Kunlun Tech to build SkyReels AI, an AI short-drama platform, from scratch.
- 2023.03Talk at the GDC Machine Learning Summit on GPT-3 powered text to lifelike speech and animation for NPCs.GDC
- 2020.09Talk at the Beijing International Game Innovation Conference on AI-assisted game content production.BIGC
- 2019.12Talk at NVIDIA GTC China on video-guided intelligent choreography, shipped in Justice Online.GTC
Show all (14)Show less
Work
From my paper-to-video pipeline: feed it an arXiv paper, get back a narrated video with shot planning and animation.
Selected papers
All papers →IEEE TPAMI 2025
VolGen: Volumetric Latent Diffusion Models for 3D Object Generation
J Tang, X Wen, HX Guo, H Jiang, Z Li, J Xu, G Zeng
ICML 2025
Policy Regularization on Globally Accessible States in Cross-Dynamics Reinforcement Learning
Z Xue, L Feng, J Xu, K Kang, X Wen, B An, S Yan
SIGGRAPH Asia 2022
AgileAvatar: Stylized 3D Avatar Creation via Cascaded Domain Bridging
S Sang, T Zhi, G Song, M Liu, C Lai, J Liu, X Wen, J Davis, L Luo
IEEE TMM 2022First author
Multi-source Multi-label Learning for User Profiling in Online Games
X Wen, S Zhao, H Wang, R Wu, M Qu, T Hu, G Chen, J Tao, C Fan
ACM MM 2023
Speech-Driven 3D Face Animation with Composite and Regional Facial Movements
H Wu, S Zhou, J Jia, J Xing, Q Wen, X Wen
arXiv 2025
SkyReels-A1: Expressive Portrait Animation in Video Diffusion Transformers
D Qiu, Z Fei, R Wang, J Bai, C Yu, M Fan, G Chen, X Wen
arXiv 2024
Video Diffusion Transformers are In-Context Learners
X Wen, et al.
FITEE 2025First author
GMCoT: A Graph-Augmented Multimodal Chain-of-Thought Reasoning Framework for Multi-label Zero-shot Learning
X Wen, H Wang, K Chen, T Hu, G Chen
arXiv 2024
MovieCharacter: A Tuning-Free Framework for Controllable Character Video Synthesis
D Qiu, Z Chen, R Wang, M Fan, C Yu, J Huang, X Wen
arXiv 2024
SkyScript-100M: 1,000,000,000 Pairs of Scripts and Shooting Scripts for Short Drama
J Tang, Q Jia, Y Xie, Z Gong, X Wen, J Zhang, Y Guo, G Chen, J Yang
arXiv 2024
InTeX: Interactive Text-to-Texture Synthesis via Unified Depth-Aware Inpainting
J Tang, R Lu, X Chen, X Wen, G Zeng, Z Liu
Computer Graphics Forum 2025
LucidFusion: Reconstructing 3D Gaussians with Arbitrary Unposed Images
H He, Y Liang, L Wang, Y Cai, X Xu, H Guo, X Wen, Y Chen
IEICE Transactions 2025First author
Semantic Web-Based Latent Alignment and Label Propagation Model for Multi-label Learning with Missing Labels
X Wen, Y Yang, H Wang, K Chen, T Hu, G Chen
arXiv 2024
Unsupervised Cross-Domain Regression for Fine-grained 3D Game Character Reconstruction
Q Wen, X Wen, H Jiang, S Yang, B Han, T Hu, G Chen, S Li
arXiv 2020
Semi-Supervised Learning for In-Game Expert-Level Music-to-Dance Translation
Y Duan, T Shi, Z Zou, J Qin, Y Zhao, Y Yuan, J Hou, X Wen, C Fan
Show 11 moreShow less
Now
Working on post-training for anime video generation at Tencent Hunyuan, focused on unified generation architectures and inference acceleration. On the side, building a pipeline that turns papers into narrated videos.
Read more →