HY-World 2.0 🔥
Core Contributor, 2026
Next-gen multi-modal world model that reconstructs, generates, and simulates 3D worlds.
News
- 2026.04HY-World 2.0 is released!
- 2025.08Three papers accepted to SIGGRAPH Asia 2025. See you in Hong Kong.
- 2025.08HY-World 1.0 is released.
- 2025.02One paper accepted to CVPR 2025.
- 2025.02One paper accepted to ICLR 2025. See you in Singapore.
- 2024.04One paper accepted to SIGGRAPH 2024. See you in Denver.
- 2023.12One paper accepted to AAAI 2024.
- 2023.04One paper accepted to SIGGRAPH 2023 (Journal).
Open-Source Projects
HY-World 1.5
Core Contributor, 2025
The first open-source world model with real-time latency, long-term memory, and interactive control.
HY-World 1.0
Core Contributor, 2025
Generate immersive, editable 3D scenes from text or a single image.
Selected Publications
*equal contribution · ^intern · †corresponding author
2026
HY-World 2.0: A Multi-Modal World Model for Reconstructing, Generating, and Simulating 3D Worlds
Technical Report, 2026
Next-gen multi-modal world model that reconstructs, generates, and simulates 3D worlds.
2025
WorldPlay: Towards Long-Term Geometric Consistency for Real-Time Interactive World Modeling
Preprint, 2025
Real-time interactive world model with long-term geometric memory.
WorldMirror: Universal 3D World Reconstruction with Any-Prior Prompting
Preprint, 2025
Universal feed-forward 3D reconstruction — any input, any output.
MoCA: Mixture-of-Components Attention for Scalable Compositional 3D Generation
Preprint, 2025
Compositional 3D scene and object generation via sparse mixture-of-components attention.
StyleSculptor: Zero-Shot Style-Controllable 3D Asset Generation with Texture-Geometry Dual Guidance
Proc. ACM SIGGRAPH Asia, 2025
Training-free 3D asset generation with joint geometry and texture style guidance.
Voyager: Long-Range and World-Consistent Video Diffusion for Explorable 3D Scene Generation
ACM TOG (Proc. SIGGRAPH 2025, Journal), 2025
Long-range 3D world exploration powered by RGB-D video diffusion.
Shape-for-Motion: Precise and Consistent Video Editing with 3D Proxy
Proc. ACM SIGGRAPH Asia, 2025
Precise video object manipulation via 3D proxies and diffusion rendering.
MAGE: Single Image to Material-Aware 3D via the Multi-View G-Buffer Estimation Model
CVPR, 2025
Single-image, material-aware 3D reconstruction via G-buffer estimation.
Phidias: A Generative Model for Creating 3D Content from Text, Image, and 3D Conditions with Reference-Augmented Diffusion
ICLR, 2025
Retrieval-augmented 3D diffusion — generate from text, image, or existing 3D assets.
2024
ThemeStation: Generating Theme-Aware 3D Assets from Few Exemplars
Proc. ACM SIGGRAPH, 2024
Generate theme-consistent 3D asset galleries from just a few exemplars.
Recasting Regional Lighting for Shadow Removal
AAAI, 2024
Shadow removal that restores degraded textures conditioned on recovered illumination.
2023
Language-based Photo Color Adjustment for Graphic Designs
ACM TOG (Proc. SIGGRAPH 2023, Journal), 2023
Language-driven photo recoloring for graphic designs.
Experience, Services & Awards
Experience
- Researcher, Tencent HY World2025.08 — present
- Research Intern, Tencent HY World2025.04 — 2025.07
- Visiting Scholar, Harvard University, VCG Group2024.10 — 2025.02
- Research Intern, Shanghai AI Laboratory, 3D AIGC Team2023.09 — 2024.10
Awards
- Outstanding Academic Performance Award, CityU2023, 2024
- Outstanding Graduate, Xiamen University2020
- Outstanding Student, Xiamen University2019, 2020
- Huang Xilie Scholarship (Top 2%)2019, 2020
- First-Class Scholarship (Top 3%)2018
Services
Reviewer · SIGGRAPH, 3DV, CVPR, ECCV, ICLR, ICML, TOG, TVCG, IJCV
Teaching Assistant · Virtual Reality Technologies and Applications (CS4188 / CS5188)