He ZHU

Master's in Robotics @ CMU 27'

I work on vision-language-action models, real-to-sim-to-real systems, and robot learning from diverse data sources.

hez2 [AT] andrew [DOT] cmu [DOT] edu

Biography Hi there, thanks for dropping by! 👋 I am a Master's student at the Robotics Institute, Carnegie Mellon University, where I have the privilege to be advised by Prof.Katerina Fragkiadaki. Prior to CMU, I obtained my B.S. degree in Computer Science from Beijing Normal University in July 2025. I am interested in Robot Learning and Multi-modal Agents. My goal is to develop machines capable of open-world spatial-temporal understanding, responding intelligently to multi-modal inputs and yielding versatile actions in the physical world. Currently, I focus on Vision-Language-Action models and Real-to-Sim-to-Real frameworks. Before I also explored various topics in 3D vision.

News

Education

Publications

Representative works are highlighted    (* denotes equal contribution)

	SkillWeaver: Agentic Exploration over Neural Interaction Skills for Scalable Robot Data Generation   He Zhu, Lusen Zhao, Kwan Man Cheng, Su Li, Katerina Fragkiadaki Conference on Robot Learning 2026 (CoRL 2026)  Project Page   Paper  Code
	4DNeX: Feed-Forward 4D Generative Modeling Made Easy   Zhaoxi Chen*, Tianqi Liu*, Long Zhuo*, Jiawei Ren, Zeng Tao, He Zhu, Fangzhou Hong, Liang Pan, Ziwei Liu Paper under review  Project Page   Paper  Code
	Multi-view Reconstruction via SfM-Guided Monocular Depth Estimation  Haoyu Guo*, He Zhu*, Sida Peng, Haotong Lin, Yunzhi Yan, Tao Xie, Wenguan Wang, Xiaowei Zhou, Hujun Bao Computer Vision and Pattern Recognition 2025 (CVPR 2025, Oral Presentation, Top 3.3%)  Project Page   Paper  Code
	SAM-Guided Graph Cut for 3D Instance Segmentation  Haoyu Guo*, He Zhu*, Sida Peng, Yuang Wang, Yujun Shen, Ruizhen Hu, Xiaowei Zhou European Conference on Computer Vision 2024 (ECCV 2024)  Project Page   Paper  Code