FSX: Message Flow Sensitivity Enhanced Structural Explainer for Graph Neural Networks

2026-01-21

AI Summary

This work proposes FSX, a novel framework for explaining graph neural networks (GNNs) that reconciles computational efficiency with accurate modeling of structural interactions to enhance explanation fidelity. FSX uniquely integrates internal message flows within the GNN with cooperative game theory over the external graph structure: it identifies critical computation paths through message-flow sensitivity analysis in a single forward pass and employs a flow-aware variant of Shapley values to quantify node contributions, thereby efficiently pinpointing predictive subgraphs. Extensive experiments across multiple datasets and GNN architectures demonstrate that FSX significantly outperforms existing methods, achieving both high computational efficiency and superior explanation fidelity. Moreover, the framework reveals an intrinsic mechanism by which substructures influence predictions through modulating the stability of message propagation.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Reasoning under Uncertainty: Graphical Models
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Social Networks and Social Media: Influence propagation, information diffusion, and the prediction on networks

Abstract

Despite the widespread success of Graph Neural Networks (GNNs), understanding the reasons behind their specific predictions remains challenging. Existing explainability methods face a trade-off that gradient-based approaches are computationally efficient but often ignore structural interactions, while game-theoretic techniques capture interactions at the cost of high computational overhead and potential deviation from the model's true reasoning path. To address this gap, we propose FSX (Message Flow Sensitivity Enhanced Structural Explainer), a novel hybrid framework that synergistically combines the internal message flows of the model with a cooperative game approach applied to the external graph data. FSX first identifies critical message flows via a novel flow-sensitivity analysis: during a single forward pass, it simulates localized node perturbations and measures the resulting changes in message flow intensities. These sensitivity-ranked flows are then projected onto the input graph to define compact, semantically meaningful subgraphs. Within each subgraph, a flow-aware cooperative game is conducted, where node contributions are evaluated fairly through a Shapley-like value that incorporates both node-feature importance and their roles in sustaining or destabilizing the identified critical flows. Extensive evaluation across multiple datasets and GNN architectures demonstrates that FSX achieves superior explanation fidelity with significantly reduced runtime, while providing unprecedented insights into the structural logic underlying model predictions--specifically, how important sub-structures exert influence by governing the stability of key internal computational pathways.

Problem

• Graph Neural Networks
• Explainability
• Message Flow
• Structural Interaction
• Computational Efficiency

Innovation

• message flow sensitivity
• structural explanation
• graph neural networks
• cooperative game theory
• Shapley value