---
title: Explainable Graph Neural Networks via Structural Externalities
url: https://www.ml-quant.com/papers/arxiv/2507.17848/
site: ML-Quant (https://www.ml-quant.com)
updated: 2026-09-26
license: Summaries CC BY 4.0; links go to the original sources
index: https://www.ml-quant.com/llms.txt
identifier: arXiv:2507.17848
source_url: http://arxiv.org/abs/2507.17848v1
featured: 2025-07-25
citations: 4
topic: ML & AI Methods
---


# Explainable Graph Neural Networks via Structural Externalities

GraphEXT, a new explainability framework for Graph Neural Networks, improves their explainability by focusing on node interactions and the effect of structural changes on predictions.

- Source: http://arxiv.org/abs/2507.17848v1
- Identifier: arXiv:2507.17848
- Released: 2025-07-19
- First featured: Quant Letter No. 107 (2025-07-25): https://www.ml-quant.com/issues/2025-07-25/
- Citations (Semantic Scholar): 4
- Published in: International Joint Conference on Artificial Intelligence
- Topic: ML & AI Methods

## Related

- [Graph Mamba: Towards Learning on Graphs with State Space Models](https://www.ml-quant.com/papers/arxiv/2402.08678/): Graph Mamba Networks, a new type of Graph Neural Networks, have been introduced, which achieve excellent performance in various benchmark datasets despite lower computational cost.
- [Edge Directionality Improves Learning on Heterophilic Graphs](https://www.ml-quant.com/papers/arxiv/2305.10498/): The study presents Directed Graph Neural Network (Dir-GNN), a new deep learning framework for directed graphs that surpasses traditional models in heterophilic benchmarks.
- [A Copula-Guided In-Model Interpretable Neural Network for Change Detection in Heterogeneous Remote Sensing Images](https://www.ml-quant.com/papers/arxiv/2303.17448/): NN-Copula-CD, a new method for detecting changes in heterogeneous remote sensing images, combines copula theory and deep neural networks, improving disaster monitoring and land-use management.
- [Decoding Quantum LDPC Codes Using Graph Neural Networks](https://www.ml-quant.com/papers/arxiv/2408.05170/): A new decoding method for Quantum Low-Density Parity-Check codes based on Graph Neural Networks outperforms conventional and neural-enhanced decoding algorithms in performance and complexity.
- [GraphCNNpred: A stock market indices prediction using a Graph based deep learning system](https://www.ml-quant.com/papers/arxiv/2407.03760/): Stock Prediction: The paper introduces a graph neural network-based convolutional neural network model for predicting stock market prices, using custom feature engineering on diverse data sources.
- [GNN-VPA: A Variance-Preserving Aggregation Strategy for Graph Neural Networks](https://www.ml-quant.com/papers/arxiv/2403.04747/): A new variance-preserving aggregation function for Graph Neural Networks is proposed, which could lead to more efficient and self-normalizing networks.