Zanxin Chen (陈攒鑫)

My research interests mainly focus on Embodied AI, especially robotic foundation models, manipulation policy learning (VLA and diffusion policy), and robot data synthesis.

Portrait of Zanxin Chen

Biography

I am an undergraduate student at Shenzhen University, majoring in Mathematical Sciences and minoring in Computer Science. I am currently an intern researcher at Shanghai AI Laboratory, where I work on embodied AI and robotic manipulation. I will start my joint Ph.D. program at Shanghai Jiao Tong University and Shanghai AI Laboratory in June 2026.

Education

Shenzhen University

Shenzhen University (SZU), China

B.S. in Mathematical Sciences (currently studying)

B.Eng. minor in Computer Science (currently studying)

Sep. 2022 - Jun. 2026 (expected)

Publications | [Google Scholar]

Recent Research Highlights

RoboTwin 2.0, arXiv 2025
RoboTwin, CVPR 2025 Highlight & ECCV WS Best Paper

Full List

RoboTwin 2.0: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulation GitHub repo stars

Tianxing Chen*, Zanxin Chen*, Baijun Chen*, Zijian Cai*, Yibin Liu*, Qiwei Liang, Zixuan Li, Xianliang Lin, Yiheng Ge, Zhenyu Gu, Weiliang Deng, Yubin Guo, Tian Nian, Xuanbing Xie, Qiangyu Chen, Kailun Su, Tianling Xu, Guodong Liu, Mengkang Hu, Huan-ang Gao, Kaixuan Wang, Zhixuan Liang, Yusen Qin, Xiaokang Yang, Ping Luo†, Yao Mu†

Under Review

Project Page / Paper / arXiv / Code / Document / 具身智能之心

RoboTwin project preview

RoboTwin: Dual-Arm Robot Benchmark with Generative Digital Twins

Yao Mu*, Tianxing Chen*, Zanxin Chen*, Shijia Peng*, Zeyu Gao, Zhixuan Liang, Qiaojun Yu, Yude Zou, Mingkun Xu, Lunkai Lin, Zhiqiang Xie, Mingyu Ding and Ping Luo†

Final Version: CVPR 2025 (Highlight)

Early Version: ECCV @ MAAS 2024 (Best Paper)

Project Page / Paper / arXiv / Code / 具身智能之心

Using the COBOT Magic platform, we have collected diverse data on tool usage, human-robot interaction, and mobile manipulation. We present a cost-effective approach to creating digital twins using AI-generated content, transforming 2D images into detailed 3D models.

G3Flow project preview

G3Flow: Generative 3D Semantic Flow for Pose-aware and Generalizable Object Manipulation

Tianxing Chen*, Yao Mu*, Zhixuan Liang*, Zanxin Chen, Shijia Peng, Qiangyu Chen, Mingkun Xu, Ruizhen Hu, Hongyuan Zhang, Xuelong Li, Ping Luo†

CVPR 2025

Project Page / Paper / arXiv / Code

We present G3Flow, a novel approach that leverages foundation models to generate and maintain 3D semantic flow for enhanced robotic manipulation.

SAM2-based tracking project preview

Articulated Object Manipulation using Online Axis Estimation with SAM2-Based Tracking

Xi Wang*, Tianxing Chen*, Qiaojun Yu*, Tianling Xu, Zanxin Chen, Yiting Fu, Cewu Lu†, Yao Mu†, Ping Luo†

Under Review 2024

Project Page / Paper / arXiv / Code

This study introduces a closed-loop pipeline that integrates interactive perception with online axis estimation from segmented 3D point clouds using Segment Anything Model 2.

* Equal contribution. † Corresponding author.

Research & Visiting Experience

Shanghai AI Laboratory

Shanghai AI Laboratory (Pujiang National Laboratory)

Feb. 2024 - Present, Intern Researcher, Shanghai, China

Mentored by MuYao and PangJiangmiao

Shanghai Jiao Tong University

ScaleLab, Shanghai Jiao Tong University (SJTU)

Visiting Student, Shanghai, China

Mentored by MuYao

Honors

  • [2024] Liyuan Star Scholarship, the highest honor for college-level undergraduates, CNY 20,000
  • [2024] First Prize of Innovation & Entrepreneurship Star (Team), CNY 3,000
  • [2023] First Prize of Innovation & Entrepreneurship Star (Individual), CNY 3,000
  • [2023] First Prize of Study Star, CNY 3,000