mlsanityData to 14 Oct 2026

SentiNet: Detecting Localized Universal Attacks Against Deep Learning Systems

arXiv 1812.00292 ↗First posted 2 Dec 2018cs.CRcited by 152

Not on the map. Why?

Cited by 24 in LLM safety and security, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

SentiNet is a novel detection framework for localized universal attacks on neural networks. These attacks restrict adversarial noise to contiguous portions of an image and are reusable with different images -- constraints that prove useful for generating physically-realizable attacks. Unlike most other works on adversarial detection, SentiNet does not require training a model or preknowledge of an attack prior to detection. Our approach is appealing due to the large number of possible mechanisms and attack-vectors that an attack-specific defense would have to consider. By leveraging the neural network's susceptibility to attacks and by using techniques from model interpretability and object detection as detection mechanisms, SentiNet turns a weakness of a model into a strength. We demonstrate the effectiveness of SentiNet on three different attacks -- i.e., data poisoning attacks, trojaned networks, and adversarial patches (including physically realizable attacks) -- and show that our defense is able to achieve very competitive performance metrics for all three threats. Finally, we show that SentiNet is robust against strong adaptive adversaries, who build adversarial patches that specifically target the components of SentiNet's architecture.

Key facts. Cited by 152 papers in the map (data through 14 Oct 2026). First posted on arXiv 2 Dec 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 25; 2022: 33; 2023: 36; 2024: 29; 2025: 20; 2026: 9.

152 papers.

Sort:

By area:

  1. 2021Adversarial Unlearning of Backdoors via Implicit Hypergradientcited by 166
  2. 2021Rethinking the Backdoor Attacks' Triggers: A Frequency Perspectivecited by 161
  3. 2021BadEncoder: Backdoor Attacks to Pre-trained Encoders in Self-Supervised Learningcited by 123
  4. 2021RAP: Robustness-Aware Perturbations for Defending against Backdoor Attacks on NLP Modelscited by 104
  5. 2021Backdoor Scanning for Deep Neural Networks through K-Arm Optimizationcited by 98
  6. 2022Wild Patterns Reloaded: A Survey of Machine Learning Security against Training Data Poisoningcited by 94
  7. 2022Untargeted Backdoor Watermark: Towards Harmless and Stealthy Dataset Copyright Protectioncited by 86
  8. 2021Black-box Detection of Backdoor Attacks with Limited Information and Datacited by 73
  9. 2021PatchCleanser: Certifiably Robust Defense against Adversarial Patches for Any Image Classifiercited by 65
  10. 2022Backdoor Defense with Machine Unlearningcited by 65
  11. 2022The "Beatrix'' Resurrections: Robust Backdoor Detection via Gram Matricescited by 63
  12. 2022MM-BD: Post-Training Detection of Backdoor Attacks with Arbitrary Backdoor Pattern Types Using a Maximum Margin Statisticcited by 55
38 more
  1. 2023UNICORN: A Unified Backdoor Trigger Inversion Frameworkcited by 46
  2. 2021An Overview of Backdoor Attacks Against Deep Neural Networks and Possible Defencescited by 42
  3. 2022Towards A Proactive ML Approach for Detecting Backdoor Poison Samplescited by 40
  4. 2021Trigger Hunting with a Topological Prior for Trojan Detectioncited by 39
  5. 2022Rethinking the Reverse-engineering of Trojan Triggerscited by 39
  6. 2022Training with More Confidence: Mitigating Injected and Natural Backdoors During Trainingcited by 38
  7. 2023Understanding the Robustness of 3D Object Detection with Bird's-Eye-View Representations in Autonomous Drivingcited by 37
  8. 2021TnT Attacks! Universal Naturalistic Adversarial Patches Against Deep Neural Network Systemscited by 36
  9. 2025Backdoor Cleaning without External Guidance in MLLM Fine-tuningcited by 35
  10. 2021Towards Practical Deployment-Stage Backdoor Attack on Deep Neural Networkscited by 34
  11. 2022On the Real-World Adversarial Robustness of Real-Time Semantic Segmentation Models for Autonomous Drivingcited by 30
  12. 2023ASSET: Robust Backdoor Data Detection Across a Multiplicity of Deep Learning Paradigmscited by 29
  13. 2023Label Poisoning is All You Needcited by 29
  14. 2024Breaking the False Sense of Security in Backdoor Defense through Re-Activation Attackcited by 28
  15. 2022Architectural Backdoors in Neural Networkscited by 25
  16. 2023TPatch: A Triggered Physical Adversarial Patchcited by 22
  17. 2022MOVE: Effective and Harmless Ownership Verification via Embedded External Featurescited by 21
  18. 2021Robust Backdoor Attacks against Deep Neural Networks in Real Physical Worldcited by 20
  19. 2023Distilling Cognitive Backdoor Patterns within an Imagecited by 20
  20. 2023Defending Against Patch-based Backdoor Attacks on Self-Supervised Learningcited by 20
  21. 2023Backdoor Attack with Sparse and Invisible Triggercited by 20
  22. 2022Expose Backdoors on the Way: A Feature-Based Efficient Defense against Textual Backdoor Attackscited by 19
  23. 2024BackdoorBench: A Comprehensive Benchmark and Analysis of Backdoor Learningcited by 19
  24. 2022Eliminating Backdoor Triggers for Deep Neural Networks Using Attention Relation Graph Distillationcited by 18
  25. 2022REAP: A Large-Scale Realistic Adversarial Patch Benchmarkcited by 18
  26. 2023UMD: Unsupervised Model Detection for X2X Backdoor Attackscited by 18
  27. 2022A Survey of Neural Trojan Attacks and Defenses in Deep Learningcited by 17
  28. 2022Adversarial Patch Attacks and Defences in Vision-Based Tasks: A Surveycited by 17
  29. 2022Universal Backdoor Attacks Detection via Adaptive Adversarial Probecited by 16
  30. 2021Enhancing Backdoor Attacks with Multi-Level MMD Regularizationcited by 15
  31. 2022Few-shot Backdoor Attacks via Neural Tangent Kernelscited by 15
  32. 2023BEAGLE: Forensics of Deep Learning Backdoor Attack for Better Defensecited by 15
  33. 2023DIFFender: Diffusion-Based Adversarial Defense against Patch Attackscited by 15
  34. 2022Jigsaw Puzzle: Selective Backdoor Attack to Subvert Malware Classifierscited by 14
  35. 2023Defenses in Adversarial Machine Learning: A Surveycited by 14
  36. 2024Preference Poisoning Attacks on Reward Model Learningcited by 14
  37. 2021TOP: Backdoor Detection in Neural Networks via Transferability of Perturbationcited by 13
  38. 2021Jujutsu: A Two-stage Defense against Adversarial Patch Attacks on Deep Neural Networkscited by 13

50 most cited of 152 All 152 (JSON)