mlsanityData to 14 Oct 2026

Intruding with Words: Towards Understanding Graph Injection Attacks at the Text Level

arXiv 2405.16405 ↗First posted 26 May 2024cs.LG cs.AIcited by 9

Not on the map. Why?

Abstract (arXiv)

Graph Neural Networks (GNNs) excel across various applications but remain vulnerable to adversarial attacks, particularly Graph Injection Attacks (GIAs), which inject malicious nodes into the original graph and pose realistic threats. Text-attributed graphs (TAGs), where nodes are associated with textual features, are crucial due to their prevalence in real-world applications and are commonly used to evaluate these vulnerabilities. However, existing research only focuses on embedding-level GIAs, which inject node embeddings rather than actual textual content, limiting their applicability and simplifying detection. In this paper, we pioneer the exploration of GIAs at the text level, presenting three novel attack designs that inject textual content into the graph. Through theoretical and empirical analysis, we demonstrate that text interpretability, a factor previously overlooked at the embedding level, plays a crucial role in attack strength. Among the designs we investigate, the Word-frequency-based Text-level GIA (WTGIA) is particularly notable for its balance between performance and interpretability. Despite the success of WTGIA, we discover that defenders can easily enhance their defenses with customized text embedding methods or large language model (LLM)--based predictors. These insights underscore the necessity for further research into the potential and practical significance of text-level GIAs.

Key facts. Cited by 9 papers in the map (data through 14 Oct 2026). Builds on 29 of its 38 references that are in the map. First posted on arXiv 26 May 2024.

Builds on

29 of 38 references.

References · 29

  1. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  2. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
  3. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
  4. 2017Graph Attention Networkscited by 8,923
  5. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  6. 2017Inductive Representation Learning on Large Graphscited by 6,320
  7. 2020Open Graph Benchmark: Datasets for Machine Learning on Graphscited by 2,163
  8. 2021Large Dual Encoders Are Generalizable Retrieverscited by 456
  9. 2018Adversarial Attacks on Neural Networks for Graph Datacited by 397
  10. 2023Exploring the Potential of Large Language Models (LLMs) in Learning on Graphscited by 274
  11. 2019Adversarial Attacks on Graph Neural Networks via Meta Learningcited by 261
  12. 2023Language is All a Graph Needscited by 220
17 more
  1. 2023GPT4Graph: Can Large Language Models Understand Graph Structured Data ? An Empirical Evaluation and Benchmarkingcited by 210
  2. 2019Topology Attack and Defense for Graph Neural Networks: An Optimization Perspectivecited by 196
  3. 2023Text Embeddings Reveal (Almost) As Much As Textcited by 192
  4. 2023Graph Foundation Models: Concepts, Opportunities and Challengescited by 103
  5. 2021Robustness of Graph Neural Networks at Scalecited by 100
  6. 2023A Survey of Graph Meets Large Language Model: Progress and Future Directionscited by 85
  7. 2024Can GNN be Good Adapter for LLMs?cited by 71
  8. 2021TDGIA:Effective Injection Attacks on Graph Neural Networkscited by 67
  9. 2022Understanding and Improving Graph Injection Attack by Promoting Unnoticeabilitycited by 62
  10. 2023Graph Prompt Learning: A Comprehensive Survey and Beyondcited by 61
  11. 2020Scalable Attack on Graph Data by Injecting Vicious Nodescited by 50
  12. 2021Graph Robustness Benchmark: Benchmarking the Adversarial Robustness of Graph Machine Learningcited by 44
  13. 2018Attack Graph Convolutional Networks by Adding Fake Nodescited by 26
  14. 2024A survey of dynamic graph neural networkscited by 24
  15. 2021How does Heterophily Impact the Robustness of Graph Neural Networks? Theoretical Connections and Practical Implicationscited by 21
  16. 2022GANI: Global Attacks on Graph Neural Networks via Imperceptible Node Injectionscited by 10
  17. 2019Gradient projection and conditional gradient methods for constrained nonconvex minimizationcited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 5; 2026: 4.

9 papers.

  1. 2025Data-centric Federated Graph Learning with Large Language Modelscited by 3
  2. 2025Robustness in Text-Attributed Graph Learning: Insights, Trade-offs, and New Defensescited by 1
  3. 2025Navigating the Black Box: Leveraging LLMs for Effective Text-Level Graph Injection Attackscited by 0
  4. 2025Unveiling the Vulnerability of Graph-LLMs: An Interpretable Multi-Dimensional Adversarial Attack on TAGscited by 0
  5. 2025GRAPHTEXTACK: A Realistic Black-Box Node Injection Attack on LLM-Enhanced GNNscited by 0
  6. 2026Single-Edge Node Injection Threats to GNN-Based Security Monitoring in Industrial Graph Systemscited by 0
  7. 2026Graph-Aware Stealthy Poison-Text Backdoors for Text-Attributed Graphscited by 0
  8. 2026Can LLMs Fool Graph Learning? Exploring Universal Adversarial Attacks on Text-Attributed Graphscited by 0
  9. 2026ZeroGAR: Benchmarking the Adversarial Robustness of Zero-Shot Graph Modelscited by 0