LangGFM: A Large Language Model Alone Can be a Powerful Graph Foundation Model

TL;DR

This paper introduces LangGFM, a large language model-based graph foundation model, and GFMBench, a comprehensive benchmark with 26 datasets, to evaluate and advance the universality of GFMs across diverse graph learning tasks.

Contribution

The paper presents a new GFM called LangGFM that leverages language models for graph learning and introduces GFMBench, a systematic benchmark to evaluate GFMs comprehensively.

Findings

LangGFM achieves state-of-the-art or comparable performance across GFMBench.

GFMBench covers diverse graph learning tasks with 26 datasets.

Revisiting textualization principles enhances GFM effectiveness.

Abstract

Graph foundation models (GFMs) have recently gained significant attention. However, the unique data processing and evaluation setups employed by different studies hinder a deeper understanding of their progress. Additionally, current research tends to focus on specific subsets of graph learning tasks, such as structural tasks, node-level tasks, or classification tasks. As a result, they often incorporate specialized modules tailored to particular task types, losing their applicability to other graph learning tasks and contradicting the original intent of foundation models to be universal. Therefore, to enhance consistency, coverage, and diversity across domains, tasks, and research interests within the graph learning community in the evaluation of GFMs, we propose GFMBench-a systematic and comprehensive benchmark comprising 26 datasets. Moreover, we introduce LangGFM, a novel GFM that relies entirely on large language models. By revisiting and exploring the effective graph textualization principles, as well as repurposing successful techniques from graph augmentation and graph self-supervised learning within the language space, LangGFM achieves performance on par with or exceeding the state of the art across GFMBench, which can offer us new perspectives, experiences, and baselines to drive forward the evolution of GFMs.