Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2506.04608 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908394316627968 |
|---|---|
| author | Sun, Changsheng Li, Xinke Dong, Jin Song |
| author_facet | Sun, Changsheng Li, Xinke Dong, Jin Song |
| contents | Graph Neural Networks (GNNs) are increasingly used in critical domains, where reliable explanations are vital for supporting human decision-making. However, the common practice of graph symmetrization discards directional information, leading to significant information loss and misleading explanations. Our analysis demonstrates how this practice compromises explanation fidelity. Through theoretical and empirical studies, we show that preserving directional semantics significantly improves explanation quality, ensuring more faithful insights for human decision-makers. These findings highlight the need for direction-aware GNN explainability in security-critical applications. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_04608 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Ignoring Directionality Leads to Compromised Graph Neural Network Explanations Sun, Changsheng Li, Xinke Dong, Jin Song Machine Learning Graph Neural Networks (GNNs) are increasingly used in critical domains, where reliable explanations are vital for supporting human decision-making. However, the common practice of graph symmetrization discards directional information, leading to significant information loss and misleading explanations. Our analysis demonstrates how this practice compromises explanation fidelity. Through theoretical and empirical studies, we show that preserving directional semantics significantly improves explanation quality, ensuring more faithful insights for human decision-makers. These findings highlight the need for direction-aware GNN explainability in security-critical applications. |
| title | Ignoring Directionality Leads to Compromised Graph Neural Network Explanations |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2506.04608 |