Department of Computing
The Hong Kong Polytechnic University
PQ730, Mong Man Wai Building
Hung Hom, Kowloon, Hong Kong
Wanyu is an Assistant Professor in the Department of Computing at The Hong Kong Polytechnic University. Wanyu was an Associate Editor for IEEE Transactions On Neural Networks and Learning Systems (TNNLS) in 2022-2026. Now she is serving as an Associate Editor for IEEE Computational Intelligence Magazine, Associate Editor for IEEE Transactions on Emerging Topics in Computational Intelligence (TETCI), and on the Editor Board of Memetic Computing. Wanyu was the awardee of the PolyU Young Innovative Researcher Award (YIRA) 2025, which highlighting the outstanding contributions of young faculty members at PolyU under 35.
Wanyu’s primary research interest is in 1) AI for science. We focus on developing machine learning (ML) techniques (Geometric GNNs, diffusion, etc.) to accelerate scientific simulations and design (applications in materials, drug discovery); 2) creating collaborative AI systems (multi-agent systems) with various objectives, such as robustness, efficiency, privacy and safety.
Research Team
- Shuyong Gao (Postdoctor Fellow, Previously Ph.D. and Postdoctor at Fudan University)
- Haowei Hua (Ph.D. Candidate, Previously Master Student at the University of Electronic Science and Technology of China (UESTC))
- Jiangwen Dong (Ph.D. Candidate, Previously Undergraduate at Zhejiang University)
- Zhuoran Li (Ph.D. Candidate, Previously Master at HK PolyU and Undergraduate at University of Washington-Seattle)
- Guodong Du (Ph.D. Candidate, Previously Undergraduate at UESTC)
- Zeyu An (Ph.D. Candidate, Previously Master at the University of UESTC)
- Shuhao Li (Ph.D. Candidate jointly supervised with EIT, Previously Undergraduate at Lanzhou University)
- Pengfei Zheng (Incoming Ph.D. Candidate, Previously Master at New York University and Undergraduate at University of California, Santa Barbara)
- Yixiang Chen (Incoming Ph.D. Candidate, Previously Undergraduate at HKUST)
- Mingxuan Ouyang (Mphil Candidate, Previously Master at HK PolyU)
- Minfeng Wu (Mphil Candidate, Previously Undergraduate at Xiamen University)
- Zehui Lin (Research Assistant, Previously Undergraduate at the University of Sydney)
- Xu Sun (Research Assistant, Undergraduate at HK PolyU)
- Jingwen Yang (URIS Program, Undergraduate at HK PolyU)
- Shuxiu Jia (URIS Program, Undergraduate at HK PolyU)
- Shuang Qiu (DFinTech Candidate 2024- , Thesis -- GNNs for Company Credit Evaluation)
- Yan Wong (DFinTech Candidate 2023- , Thesis -- LLMs for Trading)
- Cheung-Yee Chung (DFinTech Graduate 2022-2025, Thesis -- Trade with AI Pattern Recognition)
- Jiayuan Zhang (Research Assistant 2025-2026), a Phd candidate at Beihang University
- Chusong Zhang (Summer Research Intern 2025), Undergraduate at Zhejiang University
- Jian Chen (Postdoctor 2023-2025), an Assistant Professor at China University of Geosciences (Wuhan)
- Zhengzhong Wang (Ph.d. 2021-2024), an Assistant Professor at Xiamen University
- Jialu Li, Research Assistant (2021), a Ph.d. Candidate at Tijian University
I have been supervising Doctor of FinTech students from Business School
Alumni
We are looking for motivated Post-Doctoral Fellows and Ph.D. Students who are interested and experienced in AI for Science and Agentic AI. Send me your CV (GPA, publications, etc.) and your transcript via email if you are interested in working with me at PolyU. Candidates who have strong mathematics or physics backgrounds are preferred.
news
| May 1, 2026 |
Our four papers “Controllable Molecule Generation via Sparse Representation Editing: An Interpretability-Driven Perspective,”
“Large-Scale Molecular Dynamics Simulations: Direct Interatomic Modeling with Dilated Message Passing,”
“Discrete Diffusion with Physical Mass Constraints for De Novo Peptide Sequencing,”
and “HybridFlow: Resource-Adaptive Subtask Routing for Efficient Edge-Cloud LLM Inference,,”
were accepted by ICML 2026
|
|---|---|
| Mar 18, 2026 |
Our three papers, “MMCP-GEN: A Modality-Extensible Diffusion Language Model for Conditional Protein Sequence Generation” was accepted by CVPR 2026, “Explainable Molecular Property Prediction: Aligning Chemical Concepts with Predictions via Language Models” was accepted by TPAMI, and our survey paper “Crystalline Material Discovery in the Era of Artificial Intelligence” was accepted by ACM Computing Survey 2026.
|
| Feb 26, 2026 |
Our two papers, “Stealthy Targeted Poisoning Attacks in Vertical Split Learning via Embedding Model Manipulation” was accepted by TDSC, “The Aggregated Model is a Confounder: Enabling Deconfounded Federated Learning for OOD Generalization” was accepted by INFOCOM 2026.
|
| Jan 26, 2026 |
Our two papers, “Geometric Graph Neural Diffusion for Stable Molecular Dynamics Simulations” and “Knowledge Fusion of LLMs via Modular SkillPacks” were accepted by ICLR 2026.
|
| Nov 16, 2025 |
Our three papers “Revisiting the Canonicalization for Fast and Accurate Crystal Tensor Property
Prediction”, “Distributional Priors Guided Diffusion for Generating 3D Molecules in Low Data Regimes”
and “S-DAG: A Subject-Based Directed Acyclic Graph for Multi-Agent Heterogeneous Reasoning” were accepted by AAAI 2026. The first two are oral papers.
|
| Sep 18, 2025 |
Our two papers “Local-Global Associative Frames for Symmetry-Preserving Crystal Structure Modeling” and “Graphs Help Graphs: Multi-Agent Graph Socialized Learning were accepted by NeurIPS 2025.
|
| Jun 28, 2025 |
Our two papers, Backdoor Defense via Enhanced Splitting and Trap Isolation, was accepted by ICCV 2025 and, The Diversity Bonus: Learning from Dissimilar Clients in Personalized Federated Learning, was accepted by TNNLS 2025
|
| Jun 25, 2025 |
Wanyu was an invited speaker for the AI for Science Workshop at the Hong Kong Polytechnic University and the AI for Science and Engineering at the Young Academy of Science Summit Sub-Conference V. She gave a talk titled “Toward Interpretable Deep Learning for Molecular and Materials Science.”
|