mlsanityData to 14 Oct 2026

On the Security Risks of Knowledge Graph Reasoning

arXiv 2305.02383 ↗First posted 3 May 2023cs.CR cs.AIcited by 5

Not on the map. Why?

Abstract (arXiv)

Knowledge graph reasoning (KGR) -- answering complex logical queries over large knowledge graphs -- represents an important artificial intelligence task, entailing a range of applications (e.g., cyber threat hunting). However, despite its surging popularity, the potential security risks of KGR are largely unexplored, which is concerning, given the increasing use of such capability in security-critical domains. This work represents a solid initial step towards bridging the striking gap. We systematize the security threats to KGR according to the adversary's objectives, knowledge, and attack vectors. Further, we present ROAR, a new class of attacks that instantiate a variety of such threats. Through empirical evaluation in representative use cases (e.g., medical decision support, cyber threat hunting, and commonsense reasoning), we demonstrate that ROAR is highly effective to mislead KGR to suggest pre-defined answers for target queries, yet with negligible impact on non-target ones. Finally, we explore potential countermeasures against ROAR, including filtering of potentially poisoning knowledge and training with adversarially augmented queries, which leads to several promising research directions.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Builds on 36 of its 70 references that are in the map. First posted on arXiv 3 May 2023.

Builds on

36 of 70 references.

References · 36

  1. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
  2. 2017Graph Attention Networkscited by 8,923
  3. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  4. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  5. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,998
  6. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,268
  7. 2016Stealing Machine Learning Models via Prediction APIscited by 619
  8. 2017Wild Patterns: Ten Years After the Rise of Adversarial Machine Learningcited by 414
  9. 2018Adversarial Attacks on Neural Networks for Graph Datacited by 397
  10. 2023Poisoning Web-Scale Training Datasets is Practicalcited by 290
  11. 2019Inductive Relation Prediction by Subgraph Reasoningcited by 239
  12. 2019KagNet: Knowledge-Aware Graph Networks for Commonsense Reasoningcited by 216
24 more
  1. 2019Topology Attack and Defense for Graph Neural Networks: An Optimization Perspectivecited by 196
  2. 2020Query2box: Reasoning over Knowledge Graphs in Vector Space using Box Embeddingscited by 171
  3. 2018RSA: Byzantine-Robust Stochastic Aggregation Methods for Distributed Learning from Heterogeneous Datasetscited by 149
  4. 2020Beta Embeddings for Multi-Hop Logical Reasoning in Knowledge Graphscited by 129
  5. 2018Embedding Logical Queries on Knowledge Graphscited by 120
  6. 2020Graph Backdoorcited by 112
  7. 2018Adversarial Attacks on Node Embeddings via Graph Poisoningcited by 109
  8. 2020Complex Query Answering with Neural Link Predictorscited by 99
  9. 2022Wild Patterns Reloaded: A Survey of Machine Learning Security against Training Data Poisoningcited by 94
  10. 2020Language Generation with Multi-Hop Reasoning on Commonsense Knowledge Graphcited by 57
  11. 2019Attacking Graph-based Classification via Manipulating the Graph Structurecited by 53
  12. 2015Traversing Knowledge Graphs in Vector Spacecited by 51
  13. 2018Model-Reuse Attacks on Deep Learning Systemscited by 49
  14. 2020Learning Visual Commonsense for Robust Scene Graph Generationcited by 47
  15. 2018HOLMES: Real-time APT Detection through Correlation of Suspicious Information Flowscited by 43
  16. 2019Data Poisoning Attack against Knowledge Graph Embeddingcited by 36
  17. 2022An Embarrassingly Simple Backdoor Attack on Self-supervised Learningcited by 31
  18. 2021Generating Fake Cyber Threat Intelligence Using Transformer-Based Modelscited by 29
  19. 2019Investigating Robustness and Interpretability of Link Prediction via Adversarial Modificationscited by 25
  20. 2021Adversarial Attacks on Knowledge Graph Embeddings via Instance Attribution Methodscited by 16
  21. 2020Enabling Efficient Cyber Threat Hunting With Cyber Threat Intelligencecited by 12
  22. 2021Poisoning Knowledge Graph Embeddings via Relation Inference Patternscited by 12
  23. 2020Learning to Deceive Knowledge Graph Augmented Models via Targeted Perturbationcited by 10
  24. 2019Cyber-All-Intel: An AI for Security related Threat Intelligencecited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 3; 2025: 1; 2026: 1.

5 papers.

  1. 2025GraphRAG under Firecited by 24
  2. 2024Watch the Watcher! Backdoor Attacks on Security-Enhancing Diffusion Modelscited by 4
  3. 2024A Cognac Shot To Forget Bad Memories: Corrective Unlearning for Graph Neural Networkscited by 2
  4. 2024Resilience in Knowledge Graph Embeddingscited by 1
  5. 2026Oracle Poisoning: Corrupting Knowledge Graphs to Weaponise AI Agent Reasoningcited by 1

Similar, not cited

Close in wording, no citation either way

  1. 2025A Survey of Task-Oriented Knowledge Graph Reasoning: Status, Applications, and Prospects
  2. 2026NeuroGraph: An AI Graph-Driven Neuro-Symbolic Framework for Explainable Threat Reasoning in Advanced Manufacturing
  3. 2025TITAN: Graph-Executable Reasoning for Cyber Threat Intelligence
  4. 2025A Few Words Can Distort Graphs: Knowledge Poisoning Attacks on Graph-based Retrieval-Augmented Generation of Large Language Models
  5. 2026LogicPoison: Logical Attacks on Graph Retrieval-Augmented Generation
  6. 2026GraphRAG-LTM: A Layered Threat Modeling Framework for Attacks on Graph Retrieval-Augmented Generation
  7. 2021Membership Inference Attacks on Knowledge Graphs
  8. 2026KEPo: Knowledge Evolution Poison on Graph-based Retrieval-Augmented Generation