Tengfei Wang

HY-World 2.0: A Multi-Modal World Model for Reconstructing, Generating, and Simulating 3D Worlds Project Lead

Tencent HY World Team

Technical Report, 2026

Transforms text, images, or video into navigable 3D worlds with meshes and Gaussian Splatting.

Pathwise Test-Time Correction for Autoregressive Long Video Generation

Xunzhi Xiang*^, Zixuan Duan*, Guiyu Zhang, Haiyu Zhang^, Zhe Gao, Junta Wu, Shaofeng Zhang, Tengfei Wang^†, Qi Fan, Chunchao Guo

preprint, 2026

Fix accumulation error in long video generation without training.

WorldCompass: Reinforcement Learning for Long-Horizon World Models

Zehan Wang^, Tengfei Wang, Haiyu Zhang^, Xuhui Zuo, Junta Wu, Haoyuan Wang, Wenqiang Sun^, Zhenwei Wang, Chenjie Cao, Hengshuang Zhao, Chunchao Guo, Zhou Zhao

preprint, 2026

RL done right for world models.

FlashWorld: High-quality 3D Scene Generation within Seconds Oral

Xinyang Li^, Tengfei Wang†, Zixiao Gu^, Shengchuan Zhang, Chunchao Guo, Liujuan Cao

ICLR, 2026

Generating 3DGS scenes in 5 seconds on a single GPU.

HY-World 1.5: Interactive World Modeling with Real-Time Latency and Geometric Consistency Project Lead

Tencent HY World Team

Technical Report, 2025

Real-time world model with long-term memory and interactive control.

WorldPlay: Towards Long-Term Geometric Consistency for Real-Time Interactive World Modeling

Wenqiang Sun*^, Haiyu Zhang*^, Haoyuan Wang*^, Junta Wu, Zehang Wang, Zhenwei Wang, Yunhong Wang, Jun Zhang, Tengfei Wang†, Chunchao Guo

preprint, 2025

Real-time world model with long-term memory.

MoCA: Mixture-of-Components Attention for Scalable Compositional 3D Generation

Zhiqi Li^, Wenhuan Li^, Tengfei Wang†, Zhenwei Wang, Junta Wu, Haoyuan Wang, Yunhan Yang^, Zehuan Huang^, Yang Li, Peidong Liu, Chunchao Guo

preprint, 2025

Sparse attention for generating modular 3D assets and scenes.

WorldMirror: Universal 3D World Reconstruction with Any-Prior Prompting

Yifan Liu*^, Zhiyuan Min*^, Zhenwei Wang*, Junta Wu, Tengfei Wang†, Yixuan Yuan, Yawei Luo, Chunchao Guo

preprint, 2025

Universal and feedforward 3D reconstruction with any input and any output.

HunyuanWorld 1.0: Generating Immersive, Explorable, and Interactive 3D Worlds from Words or Pixels Project Lead

Tencent HY World Team

Technical Report, 2025

Immersive and editable 3D scene generation from images or texts.

Voyager: Long-Range and World-Consistent Video Diffusion for Explorable 3D Scene Generation

Tianyu Huang*^, Guandong Zhengwang*^, Tengfei Wang†, Yuhao Liu^, Zhenwei Wang^, Junta Wu, Jie Jiang, Hui Li, Wangmeng Zuo, Chunchao Guo

ACM Transactions on Graphics (SIGGRAPH Asia Journal), 2025

Joint RGB and depth video generation based on camera control.

Shape-for-Motion: Precise and Consistent Video Editing with 3D Proxy

Yuhao Liu^, Tengfei Wang†, Fang Liu, Zhenwei Wang^, Rynson Lau

ACM SIGGRAPH Asia, 2025

Accurate video editing with 3D proxy and diffusion rendering.

Neural LightRig: Unlocking Accurate Object Normal and Material Estimation with Multi-Light Diffusion

Zexin He*^, Tengfei Wang*, Xin Huang^, Xingang Pan, Ziwei Liu

CVPR, 2025

Accurate single-image inverse rendering with multi-light diffusion as virtual light rigs.

SAR3D: Autoregressive 3D Object Generation and Understanding via Multi-scale 3D VQVAE

Yongwei Chen, Yushi Lan, Shangchen Zhou, Tengfei Wang, Xingang Pan

CVPR, 2025

A 3D auto-regressive model for both 3D generation and understanding.

Material Anything: Generating Materials for Any 3D Object via Diffusion Highlight

Xin Huang^, Tengfei Wang†, Ziwei Liu, Qing Wang

CVPR, 2025

A diffusion model generating PBR materials for any 3D mesh.

3DTopia-XL: High-Quality 3D PBR Asset Generation via Primitive Diffusion Highlight

Zhaoxi Chen, Jiaxiang Tang, Yuhao Dong, Ziang Cao, Fangzhou Hong, Yushi Lan, Tengfei Wang, Haozhe Xie, Tong Wu, Shunsuke Saito, Liang Pan, Dahua Lin, Ziwei Liu

CVPR, 2025

3D DiT with primitive representation.

Phidias: A Generative Model for 3D Content from Text, Image, and 3D Conditions with Reference-Augmented Diffusion

Zhenwei Wang*^, Tengfei Wang*†, Zexin He^, Gerhard Hancke, Ziwei Liu, Rynson W.H. Lau

ICLR, 2025

A 3D diffusion model with RAG, supporting 3D generation from text, image, and existing 3D models.

3DTopia: Large Text-to-3D Generation Model

Fangzhou Hong, Jiaxiang Tang, Ziang Cao, Min Shi, Tong Wu, Zhaoxi Chen, Shuai Yang, Tengfei Wang, Liang Pan, Dahua Lin, Ziwei Liu

Technical Report, 2024

Fast and high-quality text-to-3D generation.

Omni6D: Large-Vocabulary 3D Object Dataset for Category-Level 6D Object Pose Estimation

Mengchen Zhang^, Tong Wu, Tai Wang, Tengfei Wang, Ziwei Liu, Dahua Lin

ECCV, 2024

A large and realistic dataset for 6D pose estimation.

LGM: Large Multi-View Gaussian Model for High-Resolution 3D Content Creation Oral Top-5 Influential

Jiaxiang Tang, Zhaoxi Chen, Xiaokang Chen, Tengfei Wang, Gang Zeng, Ziwei Liu

ECCV, 2024

Fast and high-resolution 3D generative model with 3D Gaussian splatting.

ComboVerse: Compositional 3D Assets Creation Using Spatially-Aware Diffusion Guidance

Yongwei Chen*^, Tengfei Wang*†, Tong Wu, Xingang Pan, Kui Jia, Ziwei Liu

ECCV, 2024

Image-to-3D for compositional scenes and objects.

ThemeStation: Generating Theme-Aware 3D Assets from Few Exemplars

Zhenwei Wang^, Tengfei Wang†, Gerhard Hancke, Ziwei Liu, Rynson W.H. Lau

ACM SIGGRAPH, 2024

Create a gallery of 3D assets with a consistent theme via 3D-to-3D generation.

OpenLRM: Open-Source Large Reconstruction Models Project Lead

Zexin He^ and Tengfei Wang

Technical Report, 2023

Fast image-to-3D generating in a few seconds.

Make-It-3D: High-Fidelity 3D Creation from A Single Image with Diffusion Prior Top-15 Influential

Junshu Tang, Tengfei Wang, Bo Zhang, Ting Zhang, Ran Yi, Lizhuang Ma, Dong Chen

ICCV, 2023

Generate 3D objects from a single image by distilling 3D knowledge from a 2D diffusion model.

RODIN: A Generative Model for Sculpting 3D Digital Avatars Using Diffusion Oral · 2.5%

Tengfei Wang, Bo Zhang, Ting Zhang, Shuyang Gu, Jianmin Bao, Tadas Baltrusaitis, Jingjing Shen, Dong Chen, Fang Wen, Qifeng Chen, Baining Guo

CVPR, 2023

The first 3D diffusion model for generating high-quality 3D digital avatars as neural radiance fields.

3D GAN Inversion with Facial Symmetry Prior

Fei Yin, Yong Zhang, Xuan Wang, Tengfei Wang, Xiaoyu Li, Yuan Gong, Yanbo Fan, Xiaodong Cun, Ying Shan, Cengiz Oztireli, Yujiu Yang

CVPR, 2023

High-Quality Visual Content Creation with Foundation Generative Models Best Dissertation · HM

Tengfei Wang

PhD Thesis · HKUST, 2023

HKUST CSE Best Dissertation Award · Honorable Mention.

Pretraining for Image-to-Image Translation

Tengfei Wang, Ting Zhang, Bo Zhang, Hao Ouyang, Dong Chen, Qifeng Chen, Fang Wen

technical report, 2022

A universal framework that brings pretraining power to image-to-image translation.

High-Fidelity GAN Inversion for Image Attribute Editing

Tengfei Wang, Yong Zhang, Yanbo Fan, Jue Wang, Qifeng Chen

CVPR, 2022

A novel GAN inversion framework for high-fidelity image and video editing.

Restorable Image Operators with Quasi-Invertible Networks

Hao Ouyang*, Tengfei Wang*, Qifeng Chen

AAAI, 2022

A general framework for versatile invertible image processing operators.

Dual-Camera Super-Resolution with Aligned Attention Modules Oral · 3.4%

Tengfei Wang*, Jiaxin Xie*, Wenxiu Sun, Qiong Yan, Qifeng Chen

ICCV, 2021

First real-world dual-camera super-resolution approach; new dual-camera zoom dataset.

Internal Video Inpainting by Implicit Long-range Propagation

Hao Ouyang*, Tengfei Wang*, Qifeng Chen

ICCV, 2021

Effective video inpainting for videos from arbitrary domains; no training set or pretrained models needed.

Image Inpainting with External-Internal Learning and Monochromic Bottleneck

Tengfei Wang*, Hao Ouyang*, Qifeng Chen

CVPR, 2021

A novel internal-learning method for guided image colorization.