👋 个人信息 (Profile)
I am currently a Postdoctoral Researcher / Assistant Professor at the College of Intelligence and Computing, Tianjin University. My main research interests include multimodal graph models, AI for Science (AI4Science), network representation learning, graph neural networks, and graph pre-training. I am always open to collaboration and supervision of motivated students.
我是天津大学智能与计算学部助理研究员郭翾，主要研究领域为多模态图模型、AI4Science等。 欢迎有意向的学生联系我交流探讨。
📄 Publications
Conference Journal publications are highlighted. Presentation slides and extra materials are sometimes included.
⭐ Main Publications
- Learning Heterogeneous Network Representations for Relation Prediction via Characterizing Hierarchical and Anisotropic Generation Process
- 自适应建模网络动力学的动态链路预测方法 (Dynamic Link Prediction Method for Adaptively Modeling Network Dynamics)
- Counterfactual Learning for Higher-Order Relation Prediction in Heterogeneous Information Networks
- Learning Node Representations via Sketching the Generative Process with Events Benefits Link Prediction on Heterogeneous Networks
- Representation Learning on Heterostructures via Heterogeneous Anonymous Walks
- Structure-Enhanced Graph Neural ODE Network for Temporal Link Prediction
- Learning Stochastic Equivalence based on Discrete Ricci Curvature
- Temporal Network Embedding for Link Prediction via VAE Joint Attention Mechanism
- A Survey on Role-Oriented Network Embedding
- Role-Oriented Graph Auto-Encoder Guided by Structural Information
📚 Other Publications
- Unified Network Embedding via Mutual Fusion of Communities and Roles
- HSDP: Hypergraph and structure-aware representation learning for information diffusion prediction
- Towards Robust Heterogeneous Graph Explanations under Structural Perturbations
- Towards OOD Generalization in Dynamic Graphs via Causal Invariant Learning
- TGFormer: Towards Temporal Graph Transformer with Auto-Correlation Mechanism
- HGMP: Heterogeneous Graph Multi-Task Prompt Learning
- A Survey on Temporal Interaction Graph Representation Learning: Progress, Challenges, and Opportunities
- Graph Contrastive Learning with Node-Level Accurate Difference
- Learning Accurate Neighborhood-and Self-Information for Higher-Order Relation Prediction in Heterogeneous Information Networks
- 图神经常微分方程综述 (Survey on Graph Neural Ordinary Differential Equations)
- Disentangled Representation Learning for Structural Role Discovery
- HGN2T: A Simple but Plug-and-Play Framework Extending HGNNs on Heterogeneous Temporal Graphs
- Graph Contrastive Learning via Interventional View Generation
- Network Alignment Enhanced via Modeling Heterogeneity of Anchor Nodes
- Role Discovery-Guided Network Embedding Based on Autoencoder and Attention Mechanism
- Role-Oriented Network Embedding Based on Adversarial Learning between Higher-order and Local Features
- Role-Based Network Embedding via Structural Features Reconstruction with Degree-Regularized Constraint
🎤 Presentations & Tutorials
-
Roles in Networks - Foundations, Methods and Applications
News & Updates / 新闻动态
- May 2026🎉 Paper accepted to ESWA: "Learning Heterogeneous Network Representations for Relation Prediction via Characterizing Hierarchical and Anisotropic Generation Process"论文被 ESWA 接收：Learning Heterogeneous Network Representations for Relation Prediction via Characterizing Hierarchical and Anisotropic Generation Process
- May 2026🎉 Congratulations to Liu Shilong for his paper "AgentsKG: A Hierarchical Multi-Agent Framework for Open-Domain Knowledge Graph Construction" accepted to KDD 2026!🎉 恭喜刘世龙的论文“AgentsKG: A Hierarchical Multi-Agent Framework for Open-Domain Knowledge Graph Construction”被 KDD 2026 接收！