Exploring the Potential of Large Language Models for Heterophilic Graphs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wu, Yuxia, Li, Shujie, Fang, Yuan, Shi, Chuan
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909495188258816
author Wu, Yuxia
Li, Shujie
Fang, Yuan
Shi, Chuan
author_facet Wu, Yuxia
Li, Shujie
Fang, Yuan
Shi, Chuan
contents Large language models (LLMs) have presented significant opportunities to enhance various machine learning applications, including graph neural networks (GNNs). By leveraging the vast open-world knowledge within LLMs, we can more effectively interpret and utilize textual data to better characterize heterophilic graphs, where neighboring nodes often have different labels. However, existing approaches for heterophilic graphs overlook the rich textual data associated with nodes, which could unlock deeper insights into their heterophilic contexts. In this work, we explore the potential of LLMs for modeling heterophilic graphs and propose a novel two-stage framework: LLM-enhanced edge discriminator and LLM-guided edge reweighting. In the first stage, we fine-tune the LLM to better identify homophilic and heterophilic edges based on the textual content of their nodes. In the second stage, we adaptively manage message propagation in GNNs for different edge types based on node features, structures, and heterophilic or homophilic characteristics. To cope with the computational demands when deploying LLMs in practical scenarios, we further explore model distillation techniques to fine-tune smaller, more efficient models that maintain competitive performance. Extensive experiments validate the effectiveness of our framework, demonstrating the feasibility of using LLMs to enhance node classification on heterophilic graphs.
format Preprint
id arxiv_https___arxiv_org_abs_2408_14134
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Exploring the Potential of Large Language Models for Heterophilic Graphs
Wu, Yuxia
Li, Shujie
Fang, Yuan
Shi, Chuan
Machine Learning
Artificial Intelligence
Computation and Language
Social and Information Networks
Large language models (LLMs) have presented significant opportunities to enhance various machine learning applications, including graph neural networks (GNNs). By leveraging the vast open-world knowledge within LLMs, we can more effectively interpret and utilize textual data to better characterize heterophilic graphs, where neighboring nodes often have different labels. However, existing approaches for heterophilic graphs overlook the rich textual data associated with nodes, which could unlock deeper insights into their heterophilic contexts. In this work, we explore the potential of LLMs for modeling heterophilic graphs and propose a novel two-stage framework: LLM-enhanced edge discriminator and LLM-guided edge reweighting. In the first stage, we fine-tune the LLM to better identify homophilic and heterophilic edges based on the textual content of their nodes. In the second stage, we adaptively manage message propagation in GNNs for different edge types based on node features, structures, and heterophilic or homophilic characteristics. To cope with the computational demands when deploying LLMs in practical scenarios, we further explore model distillation techniques to fine-tune smaller, more efficient models that maintain competitive performance. Extensive experiments validate the effectiveness of our framework, demonstrating the feasibility of using LLMs to enhance node classification on heterophilic graphs.
title Exploring the Potential of Large Language Models for Heterophilic Graphs
topic Machine Learning
Artificial Intelligence
Computation and Language
Social and Information Networks
url https://arxiv.org/abs/2408.14134