Wangzhi Zhan
Ph.D. Student
Department of Computer Science
Virginia Tech, Blacksburg, VA, USA
Email: wzhan24 [at] vt [dot] edu
GitHub: wzhan24
About
I am a Ph.D. student in Computer Science at Virginia Tech, advised by Prof. Dawei Zhou. My research lies at the intersection of AI for Science and geometric deep learning, with a focus on data-driven and knowledge-guided mechanical metamaterial design. I develop unified frameworks that integrate 3D topology, density conditions, and mechanical properties for efficient and controllable material generation and evaluation.
My recent work explores the use of diffusion models, symbolic reasoning, and human-in-the-loop systems to enable automated hypothesis generation, structure refinement, and interactive exploration for scientific discovery.
Research
AI for Mechanical Metamaterial Discovery
My research goal is to build a general-purpose AI system that can reason about, generate, and evaluate mechanical metamaterials from a multi-modal and multi-objective perspective. Key areas include:
- Unified generative models for inverse and forward design of mechanical properties
- 3D geometric learning and diffusion for topology generation
- Human-AI collaborative frameworks for scientific ideation
- Benchmarks and toolkits for robust and reproducible metamaterial ML evaluation
Publications
2025
UniMate: A Unified Model for Mechanical Metamaterial Generation, Property Prediction, and Condition Confirmation
Wangzhi Zhan, Jianpeng Chen, Dongqi Fu, Dawei Zhou
International Conference on Machine Learning (ICML), 2025
We propose UniMate, a unified generative model for mechanical metamaterial design that simultaneously handles 3D topology, density conditions, and mechanical property prediction.
📂 GitHub
MetamatBench: Integrating Heterogeneous Data, Computational Tools, and Visual Interface for Metamaterial Discovery
Jianpeng Chen, Wangzhi Zhan, Haohui Wang, Zian Jia, et al.
ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2025
This paper presents MetamatBench, a comprehensive benchmark for metamaterial discovery, combining five multimodal datasets, seventeen machine learning models, and a visual-interactive interface for human-AI collaboration.
📄 DOI
💻 Code
METASCIENTIST: A Human-AI Synergistic Framework for Automated Mechanical Metamaterial Design
Jingyuan Qi, Zian Jia, Minqian Liu, Wangzhi Zhan, et al.
North American Chapter of the Association for Computational Linguistics (NAACL), Demo Track, 2025
METASCIENTIST is a human-in-the-loop framework combining large language models, symbolic reasoning, and 3D structure diffusion for hypothesis-driven metamaterial generation and refinement.
📄 Paper
🔬 Demo
CV
A full curriculum vitae is available here:
📄 Download CV (PDF)
Talks
- “UniMate: Unified Learning for Mechanical Metamaterial Design”
ICML 2025, Vancouver, Canada
Contact
Email: wzhan24 [at] vt [dot] edu
Office: D&DS, Virginia Tech
GitHub: wzhan24
Google Scholar: Scholar Profile