mlsanityData to 14 Oct 2026

Graph Neural Network Explanations are Fragile

arXiv 2406.03193 ↗First posted 5 Jun 2024cs.CR cs.LGcited by 19

Not on the map. Why?

Abstract (arXiv)

Explainable Graph Neural Network (GNN) has emerged recently to foster the trust of using GNNs. Existing GNN explainers are developed from various perspectives to enhance the explanation performance. We take the first step to study GNN explainers under adversarial attack--We found that an adversary slightly perturbing graph structure can ensure GNN model makes correct predictions, but the GNN explainer yields a drastically different explanation on the perturbed graph. Specifically, we first formulate the attack problem under a practical threat model (i.e., the adversary has limited knowledge about the GNN explainer and a restricted perturbation budget). We then design two methods (i.e., one is loss-based and the other is deduction-based) to realize the attack. We evaluate our attacks on various GNN explainers and the results show these explainers are fragile.

Key facts. Cited by 19 papers in the map (data through 14 Oct 2026). Builds on 22 of its 60 references that are in the map. First posted on arXiv 5 Jun 2024.

Builds on

22 of 60 references.

References · 22

  1. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
  2. 2017Graph Attention Networkscited by 8,923
  3. 2017Inductive Representation Learning on Large Graphscited by 6,320
  4. 2018How Powerful are Graph Neural Networks?cited by 4,393
  5. 2019GNNExplainer: Generating Explanations for Graph Neural Networkscited by 737
  6. 2020Parameterized Explainer for Graph Neural Networkcited by 426
  7. 2018Adversarial Attack on Graph Structured Datacited by 301
  8. 2021On Explainability of Graph Neural Networks via Subgraph Explorationscited by 281
  9. 2019Adversarial Attacks on Graph Neural Networks via Meta Learningcited by 261
  10. 2019Explanations can be manipulated and geometry is to blamecited by 216
  11. 2020PGM-Explainer: Probabilistic Graphical Model Explanations for Graph Neural Networkscited by 211
  12. 2022Interpretable and Generalizable Graph Learning via Stochastic Attention Mechanismcited by 171
10 more
  1. 2019Explainability Techniques for Graph Convolutional Networkscited by 161
  2. 2020Interpreting Graph Neural Networks for NLP With Differentiable Edge Maskingcited by 153
  3. 2021Generative Causal Explanations for Graph Neural Networkscited by 124
  4. 2021Counterfactual Explanations Can Be Manipulatedcited by 113
  5. 2019Fooling Neural Network Interpretations via Adversarial Model Manipulationcited by 112
  6. 2020Towards More Practical Adversarial Attacks on Graph Neural Networkscited by 76
  7. 2022GStarX: Explaining Graph Neural Networks with Structure-Aware Cooperative Gamescited by 42
  8. 2022GNNInterpreter: A Probabilistic Generative Model-Level Explanation for Graph Neural Networkscited by 34
  9. 2023DEGREE: Decomposition Based Explanation For Graph Neural Networkscited by 18
  10. 2015Two samples test for discrete power-law distributionscited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 8; 2025: 4; 2026: 7.

19 papers.

Sort:

By area:

  1. 2024Protecting NeRFs' Copyright via Plug-And-Play Watermarking Base Modelcited by 18
  2. 2025A Comprehensive Survey on Self-Interpretable Neural Networkscited by 18
  3. 2024Graph Neural Network Causal Explanation via Neural Causal Modelscited by 9
  4. 2024Imaging Interiors: An Implicit Solution to Electromagnetic Inverse Scattering Problemscited by 8
  5. 2024GeometrySticker: Enabling Ownership Claim of Recolorized Neural Radiance Fieldscited by 6
  6. 2024Explainable Graph Neural Networks Under Firecited by 5
  7. 2026GNN Explanations that do not Explain and How to find Themcited by 5
  8. 2024Adapt PointFormer: 3D Point Cloud Analysis via Adapting 2D Visual Transformerscited by 4
  9. 2024Faithful Interpretation for Graph Neural Networkscited by 3
  10. 2026Inside the Unfair Judge: A Mechanistic Interpretability Account of LLM-as-Judge Biascited by 3
  11. 2025Deterministic Certification of Graph Neural Networks against Graph Poisoning Attacks with Arbitrary Perturbationscited by 2
  12. 2025Explaining GNN Explanations with Edge Gradientscited by 2
7 more
  1. 2026ATEX-CF: Attack-Informed Counterfactual Explanations for Graph Neural Networkscited by 2
  2. 2025Statistical Test for Saliency Maps of Graph Neural Networks via Selective Inferencecited by 1
  3. 2024Leveraging Local Structure for Improving Model Explanations: An Information Propagation Approachcited by 0
  4. 2026Exposing Vulnerabilities in Explanation for Time Series Classifiers via Dual-Target Attackscited by 0
  5. 2026Can LLMs Fool Graph Learning? Exploring Universal Adversarial Attacks on Text-Attributed Graphscited by 0
  6. 2026B-cos GNNs: Faithful Explanations through Dynamic Linearitycited by 0
  7. 2026GTA: Graph Theory Agent and Benchmark for Algorithmic Graph Reasoning with LLMscited by 0