Summary
Our research fields are as follows:
<Approaches>
Graph Representation Learning
- Graph Neural Networks (GNNs) for Node/Edge/Graph Embedding
- Knowledge Graph Representation/Completion/Validation/Construction
- Context-Aware Knowledge Graph Representation and Relational Learning
Large Language Models (LLMs)
- Multi-Modal & Knowledge-Enhanced Foundation Models
- Knowledge & LLM Distillation for Efficient Model Development
- Advanced Prompt Engineering: Chain-of-Thought (CoT), and Retrieval-Augmented Generation (RAG)
Synergizing LLMs and Graphs
- Text-to-Graph & Graph-to-Text Generation
- Graph-Structured Interaction for LLMs (GraphRAG, Graph-driven LLM Agents)
- Knowledge-grounded & Context-aware Response Generation with LLMs
<Applications>
Natural Language Processing (NLP)
- Question Answering, Information Retrieval & Extraction
- Document Analysis (Sentiment, Opinion, Topic, NER, Summarization)
Recommender Systems
- Knowledge-enhanced & Explainable Recommendations
- Conversational & Graph-based Recommendations
Graph Analytics and Prediction
- Node & Graph Classification Tasks
- Link Prediction and Knowledge Graph Completion
Poster

