LLMExplainer: Large Language Model based Bayesian Inference for Graph Explanation Generation

TL;DR

This paper introduces LLMExplainer, a novel approach that integrates Large Language Models as Bayesian inference modules to enhance Graph Neural Network explanations and address learning bias issues.

Contribution

It presents the first discussion of learning bias in GNN explanations and demonstrates how LLMs can serve as Bayesian inference to improve interpretability.

Findings

LLMExplainer effectively reduces learning bias in GNN explanations.

The Bayesian inference module improves explanation accuracy.

Experimental results validate the approach on synthetic and real datasets.

Abstract

Recent studies seek to provide Graph Neural Network (GNN) interpretability via multiple unsupervised learning models. Due to the scarcity of datasets, current methods easily suffer from learning bias. To solve this problem, we embed a Large Language Model (LLM) as knowledge into the GNN explanation network to avoid the learning bias problem. We inject LLM as a Bayesian Inference (BI) module to mitigate learning bias. The efficacy of the BI module has been proven both theoretically and experimentally. We conduct experiments on both synthetic and real-world datasets. The innovation of our work lies in two parts: 1. We provide a novel view of the possibility of an LLM functioning as a Bayesian inference to improve the performance of existing algorithms; 2. We are the first to discuss the learning bias issues in the GNN explanation problem.