This repository curates papers and resources on Large Language Models (LLMs) applied to graph data, serving researchers and practitioners in graph representation learning and LLM applications. It provides a structured overview of techniques and datasets for integrating LLMs with graph structures, aiming to advance graph-based reasoning and multimodal learning.

How It Works

The repository organizes research based on the survey paper "Large Language Models on Graphs: A Comprehensive Survey." It categorizes papers by how LLMs interact with graphs, including LLMs as predictors, graph-as-sequence representations, graph-empowered LLMs, and graph-aware LLM fine-tuning. This classification helps users navigate the rapidly evolving field and understand different approaches to leveraging LLMs for graph-related tasks.

Quick Start & Requirements

This is a curated list of academic papers and resources, not a software package. No installation or execution is required. Links to papers (PDFs) and code repositories are provided for each entry.

Highlighted Details

• Comprehensive categorization of LLM-on-graph research, covering pure graphs, text-attributed graphs, and text-paired graphs (e.g., molecules).
• Detailed listing of datasets and methodologies, including LLM as predictor, graph-as-sequence, graph-empowered LLMs, and graph-aware fine-tuning.
• Includes papers on various reasoning types: direct answering, heuristic reasoning, algorithmic reasoning, and temporal reasoning.
• Covers applications in diverse domains like academic networks, e-commerce, and chemistry (molecular data).

Maintenance & Community

The repository is continuously updated. Users are encouraged to contribute by opening issues or pull requests for new resources or corrections. The primary citation is provided for the survey paper.

Licensing & Compatibility

This repository itself is a collection of links and does not have a specific license. Individual papers and code repositories linked within will have their own licenses.

Limitations & Caveats

This resource is a survey and does not provide executable code or models. The rapidly evolving nature of LLMs means that some listed resources may become outdated.