I am a Research Intern at WorldEngine AI, focusing on agentic robotics and generalizable humanoid world-action models (WAM). I received my B.Eng. from Huazhong University of Science and Technology (HUST).
From Oct 2025 to Jan 2026, I was a Research Intern at PSI Lab at USC, supervised by Prof. Yue Wang, focusing on vision-language-action (VLA) models for humanoid loco-manipulation.
From Jul to Oct 2025, I was a Research Intern at The University of Texas at Austin, supervised by Prof. Qixing Huang and Prof. Georgios Pavlakos, focusing on leveraging video model priors for 3D human motion generation.
From 2024 to 2025, I was a Research Intern at Endless AI Lab at Westlake University, supervised by Prof. Yuliang Xiu, focusing on 3D human pose and shape estimation and representation.
From 2023 to 2024, I was a Research Intern at University of Utah & UCLA, supervised by Prof. Chenfanfu Jiang and Prof. Yin Yang, focusing on 3D garment generation. Before that, I explored physical simulation, debiased vision-language models and BCI (EEG analysis).
For more information, please refer to my CV.
News
📌
I am open to discussing potential research collaborations. Please contact me if you are interested! :)
2026-06-03
Ψ0 wins the Best Paper Award at a CVPR 2026 workshop!
2026-04-27
One paper: Ψ0 is accepted to RSS 2026 as an Oral!
2025-08-15
Invited talk at Qingke AI.
2025-07-24
ETCH is selected as the highlight of ICCV 2025!
2025-06-26
One paper: ETCH is accepted by ICCV 2025!
Research
* Equal contribution ⊙ Author order decided by coin flip Project lead † Corresponding author
WE-Dream: Generalizable Humanoid World-Action Models with Latent-Action Pretraining and Routed Attention
Under review at ICRA 2027
Boqian Li*⊙, Liang Heng*⊙, Weiduo Yuan*, Hongyi Jing, Songlin Wei, Jionghao Wang, Xiawei Liu, Muchen Xu, Haoyi Lin, Henghui Bao, Haonan Chang, Di Huang, Yue Wang†
Ψ0: An Open Foundation Model Towards Universal Humanoid Loco-Manipulation
RSS 2026
🏆 Oral, Best Paper Award (3D-LLM/VLA Workshop, CVPR 2026)
Songlin Wei*, Hongyi Jing*, Boqian Li*, Zhenyu Zhao*, Jiageng Mao, Zhenhao Ni, Sicheng He, Jie Liu, Xiawei Liu, Kaidi Kang, Sheng Zang, Weiduo Yuan, Marco Pavone, Di Huang, Yue Wang†
An open vision-language-action (VLA) model for humanoid loco-manipulation.
ETCH-X: Robustify Expressive Body Fitting to Clothed Humans with Composable Datasets
ECCV 2026
Xiaoben Li, Jingyi Wu, Zeyu Cai, Siyuan Yu, Boqian Li, Yuliang Xiu†
Disentangled undressing and dense fitting for scalable training, enabling expressive SMPL-X fitting to partial scans.
UP2You: Fast Reconstruction of Yourself from Unconstrained Photo Collections
ICLR 2026
Zeyu Cai, Ziyang Li, Xiaoben Li, Boqian Li, Zeyu Wang, Zhenyu Zhang†, Yuliang Xiu†
3D avatar reconstruction from unconstrained photo collections.
ETCH: Generalizing Body Fitting to Clothed Humans via Equivariant Tightness
ICCV 2025
⭐ Highlight Paper (Top 2.3%)
Boqian Li, Haiwen Feng, Zeyu Cai, Michael J. Black, Yuliang Xiu†
Clothed human body fitting via equivariant tightness vectors and sparse markers.
GarmentDreamer: 3DGS Guided Garment Synthesis with Diverse Geometry and Texture Details
3DV 2025
Boqian Li*, Xuan Li*, Ying Jiang*, Tianyi Xie, Feng Gao, Huamin Wang, Yin Yang, Chenfanfu Jiang
Template-based 3D garment synthesis guided by consistent 3DGS reference.
Experience
Research Intern — WorldEngine AI
Focus on agentic robotics and generalizable humanoid world-action models (WAM).
Research Intern — PSI Lab, University of Southern California
Advisor: Prof. Yue Wang
Focus on vision-language-action (VLA) models for humanoid loco-manipulation and real-time action chunking.
Research Intern — The University of Texas at Austin
Advisor: Prof. Qixing Huang and Prof. Georgios Pavlakos
Focus on leveraging video model priors for 3D human motion generation.
Research Intern — Endless AI Lab, Westlake University
Advisor: Prof. Yuliang Xiu
Focus on 3D Human Pose Shape Estimation and Representation.
Research Intern — University of Utah & UCLA
Advisor: Prof. Chenfanfu Jiang and Prof. Yin Yang
Focus on 3D Garment Synthesis and Diffusion based Generative Models.