Helin Xu | 徐赫临

I am a machine learning engineer at TikTok Inc. in San Jose, California. I received my M.S. in Computer Science from UC San Diego in 2024, and my B.Eng. in Automation from Tsinghua University in 2023.

My research is in embodied AI and robot learning, with an emphasis on generalizable robotic manipulation: 3D perception of articulated objects, reinforcement learning from point-cloud observations, cross-category domain generalization, and large-scale physics simulation for scalable robot training data. I have been very fortunate to be advised by Prof. Xiaolong Wang, Prof. He Wang, Prof. Li Yi, Prof. Hao Su, and Prof. Yebin Liu.

I am a lifelong learner and I aim to build things that work in the real world. I am always open to conversations. If you share similar interests or simply want to chat, feel free to reach out!

Email / Twitter / WeChat / Google Scholar / GitHub / LinkedIn

News

[2025/01] Joined TikTok Inc. as a machine learning engineer.

[2024/12] Received my M.S. in Computer Science from UC San Diego.

[2024/06] Research intern at Hillbot Inc. over the summer, working on 3D generation and simulation-ready assets.

[2023/12] Dora AI recognized in the Product Hunt Golden Kitty Awards Hall of Fame.

[2023/06] Received my B.Eng. in Automation from Tsinghua University.

[2023/03] GAPartNet selected as Highlight (top 2.5%) at CVPR 2023 with final reviews of all accepts.

[2023/02] Two papers accepted to CVPR 2023.

Research

DexArt: Benchmarking Generalizable Dexterous Manipulation with Articulated Objects
Chen Bao*, Helin Xu*, Yuzhe Qin, Xiaolong Wang†
CVPR 2023
ArXiv / Project Page / Code

We propose DexArt, a task suite of Dexterous manipulation with Articulated object using point cloud observation. We experiment with extensive benchmark methods that learn category-level manipulation policy on seen objects. We evaluate the policies’ generalizability on a collection of unseen objects, as well as their robustness to camera viewpoint change.

GAPartNet: Cross-Category Domain-Generalizable Object Perception and Manipulation via Generalizable and Actionable Parts
Haoran Geng*, Helin Xu*, Chengyang Zhao*, Chao Xu, Li Yi, Siyuan Huang, He Wang†
CVPR 2023, Highlight (top 2.5% of all submissions)
ArXiv / Project Page / Code & Dataset

We propose to learn generalizable object perception and manipulation skills via Generalizable and Actionable Parts, and present GAPartNet, a large-scale interactive dataset with rich part annotations.

Projects

Dora AI: Generating powerful websites, one prompt at a time
Project done during my machine learning engineer internship at Beijing Wuxianmali Technology Co., Ltd.
Project Page / Product Hunt / Twitter

Dora AI generates websites from a simple natural-language prompt, helping users create websites quickly without coding or design skills. I led the machine learning team, and curated front-end user interface datasets using a deep learning detection model. The product launched on Product Hunt and was ranked #1 Product of the Day, Product of the Week, and Product of the Month.

Experience

	Machine Learning Engineer, TikTok Inc. January 2025 - Present · San Jose, California
	Research Intern, Hillbot Inc. June 2024 - September 2024 · La Jolla, California Research Advisor: Prof. Hao Su
	Machine Learning Engineer Intern, Beijing Wuxianmali Technology Co., Ltd. March 2023 - October 2023 · Beijing, China Product Delivered: Dora AI
	Research Intern, UC San Diego — Wang Lab April 2022 - February 2023 · La Jolla, California Research Advisor: Prof. Xiaolong Wang
	Research Intern, Peking University — EPIC Lab May 2021 - March 2022 · Beijing, China Research Advisors: Prof. He Wang, Prof. Li Yi
	Undergraduate Research Assistant, Tsinghua University June 2020 - April 2021 · Beijing, China Research Advisor: Prof. Yebin Liu