This repository is a curated collection of research papers, datasets, and tools focused on the intersection of Large Language Models (LLMs) and graph-based techniques. It aims to bridge the gap between LLMs' natural language processing capabilities and the prevalence of graph structures in real-world applications, serving researchers and practitioners in AI, NLP, and graph machine learning.

How It Works

The collection categorizes research across various aspects of Graph-LLM integration, including foundational models, prompting strategies, specific applications (like node classification, knowledge graphs, molecular graphs), and advanced areas such as Graph Retrieval Augmented Generation (GraphRAG), planning, and multi-agent systems. It highlights papers that explore how LLMs can understand, reason over, and generate graph-structured data, or how graph structures can enhance LLM performance.

Quick Start & Requirements

This is a curated list of research papers and does not have a direct installation or execution command. Users are directed to individual paper repositories for specific code and setup instructions.

Highlighted Details

• Extensive coverage of recent (2023-2025) research, with a significant number of papers from top-tier conferences like NeurIPS, ICLR, KDD, and ACL.
• Categorization includes foundational Graph LLM models, prompting techniques (e.g., Graph Chain-of-Thought), and diverse applications.
• Features dedicated sections for specialized areas like Molecular Graphs, GraphRAG, Planning, and Multi-Agent Systems.
• Includes links to papers, code repositories, and project pages for many listed resources.

Maintenance & Community

The repository is community-driven, welcoming contributions via issues or pull requests for new relevant resources.

Licensing & Compatibility

The repository itself is a collection of links and does not have a specific license. Individual linked resources will have their own licenses.

Limitations & Caveats

This is a curated list and not a unified framework or library; users must refer to individual research papers for implementation details and potential limitations. The rapid pace of research means the collection is continuously updated.