mlsanityData to 7 Oct 2026

Robustness Inspired Graph Backdoor Defense

arXiv 2406.09836 ↗First posted 14 Jun 2024cs.LG cs.CRcited by 7

Not on the map. Why?

Abstract (arXiv)

Graph Neural Networks (GNNs) have achieved promising results in tasks such as node classification and graph classification. However, recent studies reveal that GNNs are vulnerable to backdoor attacks, posing a significant threat to their real-world adoption. Despite initial efforts to defend against specific graph backdoor attacks, there is no work on defending against various types of backdoor attacks where generated triggers have different properties. Hence, we first empirically verify that prediction variance under edge dropping is a crucial indicator for identifying poisoned nodes. With this observation, we propose using random edge dropping to detect backdoors and theoretically show that it can efficiently distinguish poisoned nodes from clean ones. Furthermore, we introduce a novel robust training strategy to efficiently counteract the impact of the triggers. Extensive experiments on real-world datasets show that our framework can effectively identify poisoned nodes, significantly degrade the attack success rate, and maintain clean accuracy when defending against various types of graph backdoor attacks with different properties.

Key facts. Cited by 7 papers in the map (data through 7 Oct 2026). Builds on 33 of its 44 references that are in the map. First posted on arXiv 14 Jun 2024.

Builds on

33 of 44 references.

References · 33

  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. 2020Open Graph Benchmark: Datasets for Machine Learning on Graphscited by 2,148
  6. 2017Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoningcited by 1,012
  7. 2020Geom-GCN: Geometric Graph Convolutional Networkscited by 821
  8. 2019Graph Neural Networks for Social Recommendationcited by 711
  9. 2018Detecting Backdoor Attacks on Deep Neural Networks by Activation Clusteringcited by 429
  10. 2020Backdoor Learning: A Surveycited by 412
  11. 2020Reflection Backdoor: A Natural Backdoor Attack on Deep Neural Networkscited by 341
  12. 2020Invisible Backdoor Attack with Sample-Specific Triggerscited by 338
21 more
  1. 2020Weight Poisoning Attacks on Pre-trained Modelscited by 318
  2. 2021Anti-Backdoor Learning: Training Clean Models on Poisoned Datacited by 258
  3. 2020GNNGuard: Defending Graph Neural Networks against Adversarial Attackscited by 195
  4. 2021Is Homophily a Necessity for Graph Neural Networks?cited by 182
  5. 2019Universal Litmus Patterns: Revealing Backdoor Attacks in CNNscited by 143
  6. 2020Backdoor Attacks to Graph Neural Networkscited by 126
  7. 2022A Comprehensive Survey on Trustworthy Graph Neural Networks: Privacy, Robustness, Fairness, and Explainabilitycited by 118
  8. 2020RAB: Provable Robustness Against Backdoor Attackscited by 115
  9. 2020Graph Backdoorcited by 112
  10. 2020An Embarrassingly Simple Approach for Trojan Attack in Deep Neural Networkscited by 92
  11. 2019Molecular geometry prediction using a deep generative graph neural networkcited by 91
  12. 2021CRFL: Certifiably Robust Federated Learning against Backdoor Attackscited by 91
  13. 2020On Certifying Robustness against Backdoor Attacks via Randomized Smoothingcited by 74
  14. 2022Towards Robust Graph Neural Networks for Noisy Graphs with Sparse Labelscited by 63
  15. 2020Certified Robustness of Graph Neural Networks against Adversarial Structural Perturbationcited by 57
  16. 2021Federated Social Recommendation with Graph Neural Networkcited by 56
  17. 2023Unnoticeable Backdoor Attacks on Graph Neural Networkscited by 45
  18. 2024Rethinking Graph Backdoor Attacks: A Distribution-Preserving Perspectivecited by 22
  19. 2021Label-Wise Graph Convolutional Network for Heterophilic Graphscited by 17
  20. 2023A clean-label graph backdoor attack method in node classification taskcited by 10
  21. 2023Efficient Backdoor Attacks for Deep Neural Networks in Real-world Scenarioscited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 2; 2026: 5.

7 papers.

  1. 2026Attack by Unlearning: Unlearning-Induced Adversarial Attacks on Graph Neural Networkscited by 1
  2. 2025HeteroBA: A Structure-Manipulating Backdoor Attack on Heterogeneous Graphscited by 0
  3. 2025Cross-Paradigm Graph Backdoor Attacks with Promptable Subgraph Triggerscited by 0
  4. 2026Poisoning the Inner Prediction Logic of Graph Neural Networks for Clean-Label Backdoor Attackscited by 0
  5. 2026Trapping Attacker in Dilemma: Examining Internal Correlations and External Influences of Trigger for Defending GNN Backdoorscited by 0
  6. 2026Rethinking Molecular Graph Backdoors under Chemistry-aware Admissioncited by 0
  7. 2026FedLSG: LLM-Enhanced Semantic Calibration for Federated Graph Backdoor Defensecited by 0