mlsanityData to 7 Oct 2026

TDGIA:Effective Injection Attacks on Graph Neural Networks

arXiv 2106.06663 ↗First posted 12 Jun 2021cs.LG cs.CRcited by 67

Not on the map. Why?

Abstract (arXiv)

Graph Neural Networks (GNNs) have achieved promising performance in various real-world applications. However, recent studies have shown that GNNs are vulnerable to adversarial attacks. In this paper, we study a recently-introduced realistic attack scenario on graphs -- graph injection attack (GIA). In the GIA scenario, the adversary is not able to modify the existing link structure and node attributes of the input graph, instead the attack is performed by injecting adversarial nodes into it. We present an analysis on the topological vulnerability of GNNs under GIA setting, based on which we propose the Topological Defective Graph Injection Attack (TDGIA) for effective injection attacks. TDGIA first introduces the topological defective edge selection strategy to choose the original nodes for connecting with the injected ones. It then designs the smooth feature optimization objective to generate the features for the injected nodes. Extensive experiments on large-scale datasets show that TDGIA can consistently and significantly outperform various attack baselines in attacking dozens of defense GNN models. Notably, the performance drop on target GNNs resultant from TDGIA is more than double the damage brought by the best attack solution among hundreds of submissions on KDD-CUP 2020.

Key facts. Cited by 67 papers in the map (data through 7 Oct 2026). Builds on 20 of its 27 references that are in the map. First posted on arXiv 12 Jun 2021.

Builds on

20 of 27 references.

References · 20

  1. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  2. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  3. 2017Inductive Representation Learning on Large Graphscited by 6,284
  4. 2016Layer Normalizationcited by 5,579
  5. 2013Intriguing properties of neural networkscited by 5,114
  6. 2018How Powerful are Graph Neural Networks?cited by 4,355
  7. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
  8. 2020Open Graph Benchmark: Datasets for Machine Learning on Graphscited by 2,148
  9. 2019Simplifying Graph Convolutional Networkscited by 1,470
  10. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,259
  11. 2018Graph Convolutional Neural Networks for Web-Scale Recommender Systemscited by 1,133
  12. 2018Predict then Propagate: Graph Neural Networks meet Personalized PageRankcited by 987
8 more
  1. 2018Adversarial Attacks on Neural Networks for Graph Datacited by 395
  2. 2018TextBugger: Generating Adversarial Text Against Real-world Applicationscited by 376
  3. 2017Adversarial Attacks on Neural Network Policiescited by 314
  4. 2018Adversarial Attack on Graph Structured Datacited by 300
  5. 2018Audio Adversarial Examples: Targeted Attacks on Speech-to-Textcited by 268
  6. 2019Adversarial Attacks on Graph Neural Networks via Meta Learningcited by 261
  7. 2017Topology Adaptive Graph Convolutional Networkscited by 112
  8. 2020Scalable Attack on Graph Data by Injecting Vicious Nodescited by 50

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 2; 2022: 12; 2023: 15; 2024: 13; 2025: 14; 2026: 11.

67 papers.

Sort:

By area:

  1. 2022A Comprehensive Survey on Trustworthy Graph Neural Networks: Privacy, Robustness, Fairness, and Explainabilitycited by 118
  2. 2022Trustworthy Graph Neural Networks: Aspects, Methods and Trendscited by 88
  3. 2022Understanding and Improving Graph Injection Attack by Promoting Unnoticeabilitycited by 62
  4. 2023Unnoticeable Backdoor Attacks on Graph Neural Networkscited by 45
  5. 2021Graph Robustness Benchmark: Benchmarking the Adversarial Robustness of Graph Machine Learningcited by 44
  6. 2024Graph Machine Learning in the Era of Large Language Models (LLMs)cited by 33
  7. 2022On the Robustness of Graph Neural Diffusion to Topology Perturbationscited by 30
  8. 2023Adversarial Robustness in Graph Neural Networks: A Hamiltonian Approachcited by 24
  9. 2022A Survey of Trustworthy Graph Learning: Reliability, Explainability, and Privacy Protectioncited by 20
  10. 2022Let Graph be the Go Board: Gradient-free Node Injection Attack for Graph Neural Networks via Reinforcement Learningcited by 19
  11. 2024Unleashing the Potential of Fractional Calculus in Graph Neural Networks with FRONDcited by 16
  12. 2023A Unified Framework of Graph Information Bottleneck for Robustness and Membership Privacycited by 14
38 more
  1. 2022Adversarial Camouflage for Node Injection Attack on Graphscited by 11
  2. 2022Recent Advances in Reliable Deep Graph Learning: Inherent Noise, Distribution Shift, and Adversarial Attackcited by 10
  3. 2022GANI: Global Attacks on Graph Neural Networks via Imperceptible Node Injectionscited by 10
  4. 2023Randomized Message-Interception Smoothing: Gray-box Certificates for Graph Neural Networkscited by 10
  5. 2024Coupling Graph Neural Networks with Fractional Order Continuous Dynamics: A Robustness Studycited by 9
  6. 2024Bounding the Expected Robustness of Graph Neural Networks Subject to Node Feature Attackscited by 9
  7. 2024Intruding with Words: Towards Understanding Graph Injection Attacks at the Text Levelcited by 9
  8. 2023Simple and Efficient Partial Graph Adversarial Attack: A New Perspectivecited by 8
  9. 2023Robust Graph Neural Networks via Unbiased Aggregationcited by 8
  10. 2023Node-aware Bi-smoothing: Certified Robustness against Graph Injection Attackscited by 8
  11. 2021Cluster Attack: Query-based Adversarial Attacks on Graphs with Graph-Dependent Priorscited by 7
  12. 2023Node Injection for Class-specific Network Poisoningcited by 5
  13. 2023Certifiably Robust Graph Contrastive Learningcited by 5
  14. 2024Explainable Graph Neural Networks Under Firecited by 5
  15. 2023IDEA: Invariant Defense for Graph Adversarial Robustnesscited by 4
  16. 2024Hypergraph Attacks via Injecting Homogeneous Nodes into Elite Hyperedgescited by 4
  17. 2025Exact Verification of Graph Neural Networks with Incremental Constraint Solvingcited by 4
  18. 2022Person Text-Image Matching via Text-Feature Interpretability Embedding and External Attack Node Implantationcited by 3
  19. 2023Single Node Injection Label Specificity Attack on Graph Neural Networks via Reinforcement Learningcited by 3
  20. 2025Graph Defense Diffusion Modelcited by 3
  21. 2025Query-Based and Unnoticeable Graph Injection Attack from Neighborhood Perspectivecited by 3
  22. 2023Graph Adversarial Immunization for Certifiable Robustnesscited by 2
  23. 2023Everything Perturbed All at Once: Enabling Differentiable Graph Attackscited by 2
  24. 2024Are Your Models Still Fair? Fairness Attacks on Graph Neural Networks via Node Injectionscited by 2
  25. 2025RoBCtrl: Attacking GNN-Based Social Bot Detectors via Reinforced Manipulation of Bots Control Interactioncited by 2
  26. 2026ATEX-CF: Attack-Informed Counterfactual Explanations for Graph Neural Networkscited by 2
  27. 2022Detecting Topology Attacks against Graph Neural Networkscited by 1
  28. 2022Resisting Graph Adversarial Attack via Cooperative Homophilous Augmentationcited by 1
  29. 2023GNNBleed: Inference Attacks to Unveil Private Edges in Graphs with Realistic Access to GNN Modelscited by 1
  30. 2024Understanding the Robustness of Graph Neural Networks against Adversarial Attackscited by 1
  31. 2024Adversarial Robustness of Graph Transformerscited by 1
  32. 2024RIDA: A Robust Attack Framework on Incomplete Graphscited by 1
  33. 2025LiSA: Leveraging Link Recommender to Attack Graph Neural Networks via Subgraph Injectioncited by 1
  34. 2025Adverseness vs. Equilibrium: Exploring Graph Adversarial Resilience through Dynamic Equilibriumcited by 1
  35. 2025HeTa: Relation-wise Heterogeneous Graph Foundation Attack Modelcited by 1
  36. 2025Devil's Hand: Data Poisoning Attacks to Locally Private Graph Learning Protocolscited by 1
  37. 2026Attack by Unlearning: Unlearning-Induced Adversarial Attacks on Graph Neural Networkscited by 1
  38. 2023Node Embedding from Hamiltonian Information Propagation in Graph Neural Networkscited by 0

50 most cited of 67 All 67 (JSON)