Xiaoyi Lin

Graduate Student

Wuhan University
Wuhan, China.

Email: lightcone.fake [at] gmail [dot] com

Biography

I am a first-year graduate student at Wuhan University, majoring in Artificial Intelligence at the School of Computer Science, supervised by Prof. Miao Li. I completed my undergraduate studies in Computer Science at the Hongyi Honor School with a GPA of 3.87/4.00, ranking 15th out of 129 students. I have a cross-disciplinary background, having won a bronze medal in the 37th National Physics Olympiad and a provincial first prize in the National Biology Olympiad.

My research interests include Hand-Object Interaction (HOI) tracking, dexterous manipulation, reinforcement learning, and embodied AI. I focus on advancing frameworks that bridge perception, planning, and control in robotics. I have experience in developing scalable frameworks for synthesizing manipulation trajectories and implementing reinforcement learning-based tracking controllers.

I am actively seeking PhD opportunities and collaborations.

Education

Wuhan University, China

Master of Engineering in Artificial Intelligence

School of Computer Science

Advisor: Prof. Miao Li

Sep. 2025 - Present

Wuhan University, China

Bachelor of Engineering in Computer Science

Hongyi Honor School

GPA: 3.87/4.00, Ranking: 15/129

Sep. 2021 - Jun. 2025

Internship Experience

BrainCo logo

BrainCo

Research Intern

Jun. 2026 - Present

Current research appointment.
Noitom Robotics logo

Noitom Robotics

Research Intern, Shenzhen, China

Dec. 2025 - Jun. 2026

Closely worked with Runyi Yu.

Topic: Humanoid Loco-Manipulation
Zhejiang University CAD&CG Lab logo

Zhejiang University CAD&CG Lab

Research Intern, Hangzhou, China

Jul. 2025 - Sep. 2025

Supervised by Haochen.

Topic: Hierarchical VLA system for dual-arm manipulation with grippers
NUS logo

NUS Lin's Lab

Research Intern, Singapore

Jul. 2024 - Oct. 2024

Supervised by Lin Shao.

Topic: Collision avoidance for dexterous hand manipulation

Research Papers & Projects

                                  

OmniContact: Chaining Meta-Skills via Contact Flow for Generalizable Humanoid Loco-Manipulation

Accepted to CoRL 2026

Runyi Yu*, Xiaoyi Lin*, Ji Ma, Yinhuai Wang, Koukou Luo, Jiahao Ji, Huayi Wang, Wenjia Wang, Runhan Zhang, Ping Tan, Ting Wu, Ruoli Dai, Qifeng Chen, Lei Han

[arXiv] [project page] [code] [dataset] [live demo]

DexFormer Demo

DexFormer: Learning Dexterous Manipulation via Hierarchical Transformers

Preprint Arxiv

Ke Zhang*, Lixin Xu*, Chengyi Song, Junzhe Xu, Xiaoyi Lin, Zeyu Jiang, Renjing Xu

[arXiv] [project page] [code]

HOT: Learning Generalizable Hand-Object Tracking without Human Demonstration

Accepted to NeurIPS 2026

Runyi Yu*, Xiaoyi Lin*, Hok Wai Tsui* ,Yinhuai Wang*, Hui Zhang, Qihan Zhao, Ke Fan, Miao Li, Jie Song, Jingbo Wang, Qifeng Chen, Ping Tan

[paper] [project page] [code]

GAE: Unleashing Physical Potential of VLM with Generalizable Action Expert

Accepted to ICML 2026

Mingyu Liu, Zheng Huang, Xiaoyi Lin, Muzhi Zhu, Canyu Zhao, Zongze Du, Yating Wang, Haoyi Zhu, Hao Chen, Chunhua Shen

[arXiv]

NoTVLA: Narrowing of Dense Action Trajectories for Generalizable Robot Manipulation

Accepted to NeurIPS 2026

Zheng Huang, Mingyu Liu, Xiaoyi Lin, Muzhi Zhu, Canyu Zhao, Zongze Du, Xiaoman Li, Yiduo Jia, Hao Zhong, Hao Chen, Chunhua Shen

[arXiv]

DexFlow: Unified Framework for Dexterous Manipulation

Preprint Arxiv

Xiaoyi Lin, Kunpeng Yao, Lixin Xu, Xueqiang Wang, Xuetao Li, Yuchen Wang, Miao Li

[arXiv] [project page]



Global visit map

© Xiaoyi Lin