I am a second-year Ph.D. student (2025 Fall) at the College of Computer Science and Artificial Intelligence, Fudan University, advised by Prof. Bo Dai @ HKU and Prof. Dahua Lin @ CUHK. I am currently interning at Shanghai AI Lab, working on Generative Models for Embodied AI.
My research interest includes World Model, 3D Generation and structured visual modeling.
🔥 News
- 2026.07: 🎉🎉 Our work MegaParts: Scaling Part-Aware 3D Object Generation to 300 Parts via Token-Efficient Autoregressive Modeling has been conditionally accepted to Siggraph Asia 2026. See you in Malaysia!
📝 Publications

Infinite Mobility: Scalable High-Fidelity Synthesis of Articulated Objects via Procedural Generation
Xinyu Lian, Zichao Yu, Ruiming Liang, Yitong Wang, Li Ray Luo, Kaixu Chen, Yuanzhen Zhou, Qihong Tang, Xudong Xu, Zhaoyang Lyu, Bo Dai, Jiangmiao Pang
- A procedural framework for articulated objects generation.

MeshCoder: LLM-Powered Structured Mesh Code Generation from Point Clouds
Bingquan Dai*, Li Ray Luo*, Qihong Tang, Jie Wang, Xinyu Lian, Hao Xu, Minghan Qin, Xudong Xu, Bo Dai, Haoqian Wang, Zhaoyang Lyu, Jiangmiao Pang
- Turning point cloud to shape code.

Manwen Liao*, Xinyu Lian*, Jian Mao*, Kaixu Chen, Li Luo, Jinghao Yan, Wanshui Gan, Qiao Yu, Weitian Zhang,Chunhua Shen, Guang Chen, Bo Dai, Xudong Xu, Zhaoyang Lyu
Code coming soon!
- Part-aware 3D generation to 300 parts.

From Parts to Sim: Bottom-up Autoregressive Generation of Sim-Ready 3D Assets
Xinyu Lian*, Jian Mao*, Wanshui Gan*, Jinghao Yan, Kaixu Chen, Manwen Liao, Weipeng Zhong, Qiao Yu, Yunsong Zhou, Jiafei Cao, Chunhua Shen, Guang Chen, Bo Dai, Zhaoyang Lyu
Code coming soon!
- Sim-Ready 3D assets generation.
- More works regarding artist mesh generation voxel neural renderer and scene-level real2sim are coming!
📖 Educations
- 2021.09 - 2025.06, B.E., School of Computer Science and Technology, South China University of Technology.
- 2025.09 - 2030.06(now), PhD., College of Computer Science and Artificial Intelligence, Fudan University.
💻 Internships
- 2024.10 till now, Shanghai AI Lab, Physical Intelligence Center, China.
- 2026.09 till now, JD.com, Joy Future Academy, China.