mlsanityData to 7 Oct 2026

Measuring What Matters: A Unified Evaluation Framework for GNN Explainability

arXiv 2607.04600 ↗First posted 6 Jul 2026cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Graph eXplainable AI (G-XAI) is increasingly important for making Graph Neural Networks interpretable and accountable. While a growing number of explainers are available, choosing the right method and assessing the trustworthiness of its outputs remains unclear. Consistent evaluation practices and actionable guidance are still missing, hindering practical adoption. In this paper, we introduce a unified, quantitative benchmarking framework for G-XAI that requires no ground-truth assumptions. We formalize tabular explainability metrics for graph data, evaluating topological structure and node features as independent components. Our large-scale benchmarking study identifies explainers that consistently lie on the Pareto front across metric pairs and tasks, establishing robustly non-dominated solutions - while confirming that no single explainer achieves universal superiority. We distill our findings into actionable G-XAI usability guidelines to support Machine Learning practitioners in evaluating and deploying trustworthy GNN-based pipelines.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 30 of its 43 references that are in the map. First posted on arXiv 6 Jul 2026.

Builds on

30 of 43 references.

References · 30

  1. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  2. 2017Graph Attention Networkscited by 8,854
  3. 2017Inductive Representation Learning on Large Graphscited by 6,284
  4. 2018How Powerful are Graph Neural Networks?cited by 4,355
  5. 2017Axiomatic Attribution for Deep Networkscited by 3,350
  6. 2019A Comprehensive Survey on Graph Neural Networkscited by 2,906
  7. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,348
  8. 2013Visualizing and Understanding Convolutional Networkscited by 2,278
  9. 2017Learning Important Features Through Propagating Activation Differencescited by 1,454
  10. 2014Striving for Simplicity: The All Convolutional Netcited by 983
  11. 2016Revisiting Semi-Supervised Learning with Graph Embeddingscited by 970
  12. 2019GNNExplainer: Generating Explanations for Graph Neural Networkscited by 735
18 more
  1. 2019Multi-scale Attributed Node Embeddingcited by 543
  2. 2020Parameterized Explainer for Graph Neural Networkcited by 423
  3. 2021On Explainability of Graph Neural Networks via Subgraph Explorationscited by 280
  4. 2020GraphLIME: Local Interpretable Model Explanations for Graph Neural Networkscited by 209
  5. 2020Interpreting Graph Neural Networks for NLP With Differentiable Edge Maskingcited by 152
  6. 2020Higher-Order Explanations of Graph Neural Networks via Relevant Walkscited by 138
  7. 2021CF-GNNExplainer: Counterfactual Explanations for Graph Neural Networkscited by 118
  8. 2022Evaluating Explainability for Graph Neural Networkscited by 85
  9. 2020On quantitative aspects of model interpretabilitycited by 71
  10. 2022Global Explainability of GNNs via Logic Combination of Learned Conceptscited by 64
  11. 2022GraphFramEx: Towards Systematic Evaluation of Explainability Methods for Graph Neural Networkscited by 57
  12. 2022CLEAR: Generative Counterfactual Explanations on Graphscited by 48
  13. 2023DEGREE: Decomposition Based Explanation For Graph Neural Networkscited by 18
  14. 2022BAGEL: A Benchmark for Assessing Graph Neural Network Explanationscited by 11
  15. 2023GNNX-BENCH: Unravelling the Utility of Perturbation-based GNN Explainers through In-depth Benchmarkingcited by 11
  16. 2024A Survey on Self-Supervised Graph Foundation Models: Knowledge-Based Perspectivecited by 9
  17. 2022GRAPHSHAP: Explaining Identity-Aware Graph Classifiers Through the Language of Motifscited by 6
  18. 2025B-XAIC Dataset: Benchmarking Explainable AI for Graph Neural Networks Using Chemical Datacited by 4

Built on by

Not cited yet.