About Me
I am a "Shuimu Scholar" postdoctoral researcher in the Department of Computer Science and Technology, Tsinghua University. I received my Ph.D. from the same department under the supervision of Prof. Shi-Min Hu, and my B.Eng. from Xidian University.
I have published papers in top venues such as T-PAMI, ACM TOG, SIGGRAPH, NeurIPS, ICML, ICLR and CVPR, including two CVMJ Best Paper Awards. I have received the Tsinghua University Special Scholarship (the highest honor for students at Tsinghua) and the National Scholarship. I am also a core developer of the Jittor deep learning framework. My research is supported by the Postdoctoral Innovative Talent Support Program (博新计划), Class A, and was previously supported by the inaugural NSFC Young Students Basic Research Program (Ph.D. Students). I was named in the Stanford University/Elsevier World's Top 2% Scientists list.
Citation metrics from Google Scholar, October 2026.
Join Us
I am always looking for highly self-motivated research interns, as well as prospective Ph.D. and Master's students (typically selected from interns). My current focus is embodied AI and foundation models. If you are interested in working with me, please send your CV to gmh@tsinghua.edu.cn.
News
- SkinTokens is accepted by ACM TOG.
- FreeAct is accepted by NeurIPS 2026.
- One paper is accepted by ECCV 2026.
- One paper is accepted by ICML 2026.
- GPR is accepted by KDD 2026.
- Two papers (Pixal3D and TopoCap) are accepted by SIGGRAPH 2026.
- Bee is accepted by ICLR 2026.
- One paper is published in IEEE TVCG.
- Two papers, including RBench-V, are accepted by NeurIPS 2025.
- Selected as a "Shuimu Scholar" at Tsinghua University.
- R-Bench is accepted by ICML 2025.
- UniRig is conditionally accepted by SIGGRAPH 2025 (Journal Track).
- One paper is accepted by CVPR 2025.
- Received the Tsinghua University Special Scholarship (the highest honor for students at Tsinghua, 10 recipients per year).
- Received the National Scholarship.
- One paper is accepted by T-PAMI.
Selected Publications
* equal contribution. See the full publication list or Google Scholar.
RBench-V: A Primary Assessment for Visual Reasoning Models with Multimodal Outputs
R-Bench: Graduate-level Multi-disciplinary Benchmarks for LLM & MLLM Complex Reasoning Evaluation
Bee: A High-Quality Corpus and Full-Stack Suite to Unlock Advanced Fully Open MLLMs
One Model to Rig Them All: Diverse Skeleton Rigging with UniRig
CharacterGen: Efficient 3D Character Generation from Single Images with Multi-View Pose Canonicalization
Attention Mechanisms in Computer Vision: A Survey
SegNeXt: Rethinking Convolutional Attention Design for Semantic Segmentation
Beyond Self-Attention: External Attention Using Two Linear Layers for Visual Tasks
Is Attention Better Than Matrix Decomposition?
Education & Experience
"Shuimu Scholar" Postdoctoral Researcher
Department of Computer Science and Technology, Tsinghua University
Ph.D. in Computer Science
Department of Computer Science and Technology, Tsinghua University. Advisor: Prof. Shi-Min Hu
B.Eng. in Computer Science
School of Computer Science and Technology, Xidian University. Advisor: Prof. Nannan Wang