Integrating Graphs with Large Language Models: Methods and Prospects

TL;DR

This paper reviews methods for combining large language models with graph-structured data, highlighting two main approaches: using LLMs for graph learning and employing graphs to enhance LLM capabilities, and discusses future research directions.

Contribution

It categorizes existing integration methods of LLMs and graphs, and proposes open questions to guide future research in this emerging area.

Findings

LLMs can augment graph algorithms and serve as prediction models.

Graphs can improve LLM reasoning and collaborative tasks.

Open questions for future integration strategies.

Abstract

Large language models (LLMs) such as GPT-4 have emerged as frontrunners, showcasing unparalleled prowess in diverse applications, including answering queries, code generation, and more. Parallelly, graph-structured data, an intrinsic data type, is pervasive in real-world scenarios. Merging the capabilities of LLMs with graph-structured data has been a topic of keen interest. This paper bifurcates such integrations into two predominant categories. The first leverages LLMs for graph learning, where LLMs can not only augment existing graph algorithms but also stand as prediction models for various graph tasks. Conversely, the second category underscores the pivotal role of graphs in advancing LLMs. Mirroring human cognition, we solve complex tasks by adopting graphs in either reasoning or collaboration. Integrating with such structures can significantly boost the performance of LLMs in various complicated tasks. We also discuss and propose open questions for integrating LLMs with graph-structured data for the future direction of the field.