This repository curates research papers on leveraging Large Language Models (LLMs) for graph-related tasks, serving as a comprehensive resource for researchers and practitioners in graph learning and NLP. It aims to provide an updated overview of the field, categorized by how LLMs are integrated into graph tasks, offering insights into advancements and future directions.

How It Works

The project organizes papers based on a taxonomy that categorizes LLM applications in graph tasks into roles such as "LLM as Enhancer," "LLM as Predictor," and "GNN-LLM Alignment." This structure highlights how LLMs can augment node features, directly predict graph properties, or align with Graph Neural Networks (GNNs) to capture both structural and contextual information, thereby enhancing graph learning capabilities.

Quick Start & Requirements

This repository is a curated list of research papers and does not involve direct installation or execution. All listed papers include links to their respective arXiv pages and, where available, GitHub repositories for code.

Highlighted Details

• Comprehensive categorization of LLM applications in graph tasks (Enhancer, Predictor, Alignment, Annotator, Controller, etc.).
• Links to over 100 research papers with associated code repositories and conference/journal details.
• Includes a foundational survey paper: "A Survey of Graph Meets Large Language Model: Progress and Future Directions."
• Features recent benchmarks and datasets for evaluating LLMs on graph tasks.

Maintenance & Community

The repository is actively maintained, with a call for contributions via issues or pull requests for new papers or corrections. It references other related "Awesome" lists in the graph and LLM space.

Licensing & Compatibility

The repository itself is not software and thus not subject to software licensing. The linked research papers and code repositories are governed by their respective licenses.

Limitations & Caveats

This is a curated list of research papers and does not provide executable code or pre-trained models. The rapid pace of LLM research means the list may not be exhaustive or immediately reflect the absolute latest publications.