Towards Faithful Graph Explanations with Synergistic Edge Effects via Granular Balls

2026-07-23

AI Summary

Existing graph neural network explanation methods often overlook the synergistic effects among edges, leading to inaccurate importance assessments. To address this limitation, this work proposes SeeExplainer, the first approach to introduce the granular-ball mechanism into graph explanation tasks. Specifically, it decomposes the input graph into disjoint granular balls of variable sizes through granular-ball graph refinement, constructs a structural graph to explicitly capture edge-wise synergistic relationships, and generates explanation subgraphs based on the contributions of nodes and edges within this structural graph. Notably, SeeExplainer is parameter-free and leverages a non-parametric perturbation analysis strategy. Extensive experiments across multiple graph classification benchmarks demonstrate that it significantly outperforms current state-of-the-art baselines, achieving superior accuracy, fidelity, and stability in explanations.

Abstract

Instance-level explanations aim to reveal the rationale behind a model's decisions for a specific graph. Previous methods explain graph neural networks (GNNs) by selecting important edges to induce subgraphs, where edge importance is assessed by perturbing each edge and observing changes in the model predictions. However, they often neglect the synergistic effects among edges, which are crucial for accurately characterizing edge importance. To address this issue, we propose SeeExplainer, a parameter-free explainer to interpret GNNs. Specifically, we first introduce a granular-ball graph refinement mechanism that decomposes a graph into several disjoint granular-balls with no fixed size, and utilize them as nodes to construct a structural graph. This process can better capture the synergistic effects among edges. Then, we perturb nodes and edges in the structural graph to generate explanatory subgraphs based on their respective contributions. Experiments on several graph classification datasets of different networks show that SeeExplainer outperforms state-of-the-art baselines.

Problem

• graph explanation
• synergistic edge effects
• instance-level explanation
• GNN interpretability

Innovation

• granular balls
• synergistic edge effects
• GNN explanation
• parameter-free explainer
• structural graph