Qingpo Wuwu

I am a Ph.D. student in Computer Science at Peking University, advised by Prof. Shanghang Zhang. My research interests lie at the intersection of 3D reconstruction, embodied AI, and physics-informed simulation. I aim to bridge the gap between perception and action by integrating geometric understanding into world models and vision-language-action (VLA) models, ultimately empowering embodied agents to reason about and manipulate the real-world environment.

Prior to joining the integrated M.Sc./Ph.D. program at Peking University in 2022, I started my undergraduate studies in Physics at Imperial College London in 2019 and concurrently pursued a second B.Sc. in Computer Science at Goldsmiths, University of London starting in 2020.

I am currently looking for research internship positions related to Embodied AI and world models. I am also keen to collaborate on exciting projects at the intersection of vision, physics, and robotics—feel free to contact me!

Email  /  Scholar  /  GitHub

profile photo

Education

Graduate
Peking University Peking University, Beijing, China
Integrated M.Sc./Ph.D. in Computer Science
Sep. 2022 – Present

Undergraduate
Imperial College London Imperial College London
B.Sc. in Physics
2019 – 2022 		Goldsmiths, University of London Goldsmiths, University of London
B.Sc. in Computer Science
2020 – 2025

Publications

VLA & World Model
Lift3D-VLA Lift3D-VLA: Lifting VLA Models to 3D Geometry and Dynamics-Aware Manipulation
Jiaming Liu†, Qingpo Wuwu†, Zhuoyang Liu†, Hao Chen, Nuowei Han, Yueru Jia, Chenyang Gu, Xiaoqi Li, Shanghang Zhang
TPAMI, 2026 · Under Review
Page

A 3D-aware framework that lifts geometric and dynamics priors into VLA models for robotic manipulation.
ManualVLA ManualVLA: A Unified VLA Model for Chain-of-Thought Manual Generation and Robotic Manipulation
Chenyang Gu†, Jiaming Liu†, Hao Chen†, Runzhong Huang†, Qingpo Wuwu, Zhuoyang Liu, Xiaoqi Li, Ying Li, Renrui Zhang, Peng Jia, Pheng-Ann Heng, Shanghang Zhang
CVPR, 2026
Paper / Page

WoW WoW: Towards a World Omniscient World Model Through Embodied Interaction
Xiaowei Chi, Peidong Jia, Chun-Kai Fan, Xiaozhu Ju, Weishi Mi, Kevin Zhang, Zhiyuan Qin, Wanxin Tian, Kuangzhi Ge, Hao Li, Zezhong Qian, Anthony Chen, ..., Yong Dai, Qingpo Wuwu, Chengyu Bai, ..., Yike Guo, Shanghang Zhang, Jian Tang
arXiv, 2025
Paper

WoW Eval Wow, Wo, Val! A Comprehensive Embodied World Model Evaluation Turing Test
Chun-Kai Fan, Xiaowei Chi, Xiaozhu Ju, Hao Li, Yong Bao, ..., Weishi Mi, Qingpo Wuwu, Peidong Jia, ..., Yike Guo, Shanghang Zhang, Jian Tang
arXiv, 2025
Paper

Neural Rendering
SparseStreet SparseStreet: Sparse Gaussian Splatting for Real-Time Street Scene Simulation
Qingpo Wuwu, Xiaobao Wei, Peng Chen, Nan Huang, Zhongyu Zhao, Hao Wang, Ming Lu, Ningning Ma, Shanghang Zhang
ACM ICMR, 2026

A compression framework for street scene Gaussian Splatting enabling real-time rendering.
EMD EMD: Explicit Motion Modeling for High-Quality Street Gaussian Splatting
Xiaobao Wei†, Qingpo Wuwu†, Zhongyu Zhao, Zhuangzhe Wu, Nan Huang, Ming Lu, Ningning Ma, Shanghang Zhang
ICCV, 2025
Paper / Page

Explicit motion decomposition for dynamic street scene reconstruction with 3D Gaussian Splatting.
INIF Implicit Neural Image Field for Biological Microscopy Image Compression
Gaole Dai, Rongyu Zhang†, Qingpo Wuwu†, Cheng-Ching Tseng†, Shaokang Wang†, Ming Lu, Tiejun Huang, Yu Zhou, Ali Ata Tuz, Matthias Gunzer, Jianxu Chen, Shanghang Zhang
Nature Computational Science, 2025
Paper / Page

Implicit neural representations for high-fidelity compression of large-scale biological microscopy images.
MixedGaussianAvatar MixedGaussianAvatar: Realistically and Geometrically Accurate Head Avatar via Mixed 2D-3D Gaussians
Peng Chen, Xiaobao Wei, Qingpo Wuwu, Xinyi Wang, Xingyu Xiao, Ming Lu
ACM MM, 2025
Paper

RoboArmGS RoboArmGS: High-Quality Robotic Arm Splatting via Bézier Curve Refinement
Hao Wang, Xiaobao Wei, Ying Li, Qingpo Wuwu, Dianxi Wu, Jiahao Cao, Ming Lu, Wenzhao Zheng, Shanghang Zhang
arXiv, 2025
Paper

OmniIndoor3D OmniIndoor3D: Comprehensive Indoor 3D Reconstruction
Xiaobao Wei, Xiaoan Zhang, Hao Wang, Qingpo Wuwu, Ming Lu, Wenzhao Zheng, Shanghang Zhang
arXiv, 2025
Paper

Physics-Informed Simulation
PINNsAgent PINNsAgent: Automated PDE Surrogation with Large Language Models
Qingpo Wuwu†, Chonghan Gao†, Tianyu Chen, Yihang Huang, Yuekai Zhang, Jianing Wang, Jianxin Li, Haoyi Zhou, Shanghang Zhang
ICML, 2025
Paper

An LLM-based agent framework for automating physics-informed neural network surrogation of partial differential equations.
PDEffusion PDEffusion: Solving Equations with the Physics-Grounding Diffusion Model
Chonghan Gao, Haoyi Zhou, Wenxin Gong, Qingpo Wuwu, Tianyu Chen, Qian Yu, Shanghang Zhang, Jianxin Li
IEEE BigData, 2025
Paper

Work Experience
Simplexity Robotics Research Intern, Simplexity Robotics — Embodied AI
Sep. 2025 – Present · Beijing, China
  • Developed Lift3D-VLA, a 3D-aware framework lifting geometric and dynamics priors into VLA models for robotic manipulation.
X-Humanoid Research Intern, X-Humanoid — World Model
Jul. 2025 – Sep. 2025 · Beijing, China
  • Participated in the WoW project (Towards a World Omniscient World Model through Embodied Interaction), focusing on world model development for embodied agents.
NIO Research Intern, NIO Co., Ltd. — 3D Reconstruction Division
Aug. 2024 – Present · Beijing, China
  • Developed EMD (Explicit Motion Decomposition) for dynamic scene reconstruction using Gaussian Splatting, accepted to ICCV 2025.
  • Proposed SparseStreet, a compression framework for street scene Gaussian Splatting, accepted to ACM ICMR 2026.

Awards & Grants
2026: NSFC Research Fund for Excellent International PhD Students (PI: Qingpo Wuwu) — Adaptive 3D Scene Representation Learning for Embodied AI Applications
2017: National High School Physics Competition — 13th place
2017: Provincial High School Physics Competition — 3rd place

Template from Jon Barron.