mlsanityData to 7 Oct 2026

Universal Litmus Patterns: Revealing Backdoor Attacks in CNNs

arXiv 1906.10842 ↗First posted 26 Jun 2019cs.CVcited by 143

Not on the map. Why?

Cited by 30 in LLM safety and security, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

The unprecedented success of deep neural networks in many applications has made these networks a prime target for adversarial exploitation. In this paper, we introduce a benchmark technique for detecting backdoor attacks (aka Trojan attacks) on deep convolutional neural networks (CNNs). We introduce the concept of Universal Litmus Patterns (ULPs), which enable one to reveal backdoor attacks by feeding these universal patterns to the network and analyzing the output (i.e., classifying the network as `clean' or `corrupted'). This detection is fast because it requires only a few forward passes through a CNN. We demonstrate the effectiveness of ULPs for detecting backdoor attacks on thousands of networks with different architectures trained on four benchmark datasets, namely the German Traffic Sign Recognition Benchmark (GTSRB), MNIST, CIFAR10, and Tiny-ImageNet. The codes and train/test models for this paper can be found here https://umbcvision.github.io/Universal-Litmus-Patterns/.

Key facts. Cited by 143 papers in the map (data through 7 Oct 2026). First posted on arXiv 26 Jun 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 25; 2022: 32; 2023: 33; 2024: 35; 2025: 11; 2026: 7.

143 papers.

Sort:

By area:

  1. 2024Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Trainingcited by 478
  2. 2021Neural Attention Distillation: Erasing Backdoor Triggers from Deep Neural Networkscited by 313
  3. 2021Anti-Backdoor Learning: Training Clean Models on Poisoned Datacited by 258
  4. 2021Trustworthy AI: From Principles to Practicescited by 206
  5. 2021Hidden Killer: Invisible Textual Backdoor Attacks with Syntactic Triggercited by 203
  6. 2021Advances in adversarial attacks and defenses in computer vision: A surveycited by 127
  7. 2021SPECTRE: Defending Against Backdoor Attacks Using Robust Statisticscited by 123
  8. 2024AI Agents Under Threat: A Survey of Key Security Challenges and Future Pathwayscited by 122
  9. 2021Backdoor Scanning for Deep Neural Networks through K-Arm Optimizationcited by 96
  10. 2022Wild Patterns Reloaded: A Survey of Machine Learning Security against Training Data Poisoningcited by 93
  11. 2021Backdoor Attacks on Self-Supervised Learningcited by 89
  12. 2021Turn the Combination Lock: Learnable Textual Backdoor Attacks via Word Substitutioncited by 85
38 more
  1. 2022BppAttack: Stealthy and Efficient Trojan Attacks against Deep Neural Networks via Image Quantization and Contrastive Adversarial Learningcited by 82
  2. 2023Unleashing the potential of prompt engineering for large language modelscited by 72
  3. 2021Black-box Detection of Backdoor Attacks with Limited Information and Datacited by 71
  4. 2025Poisoning Attacks on LLMs Require a Near-constant Number of Poison Samplescited by 66
  5. 2022Backdoor Defense with Machine Unlearningcited by 61
  6. 2021FIBA: Frequency-Injection based Backdoor Attack in Medical Image Analysiscited by 60
  7. 2022MM-BD: Post-Training Detection of Backdoor Attacks with Arbitrary Backdoor Pattern Types Using a Maximum Margin Statisticcited by 53
  8. 2023Detecting Backdoors in Pre-trained Encoderscited by 53
  9. 2021AEVA: Black-box Backdoor Detection Using Adversarial Extreme Value Analysiscited by 52
  10. 2021Poison Ink: Robust and Invisible Backdoor Attackcited by 50
  11. 2022A Study of the Attention Abnormality in Trojaned BERTscited by 47
  12. 2023Backdoor Defense via Adaptively Splitting Poisoned Datasetcited by 43
  13. 2021An Overview of Backdoor Attacks Against Deep Neural Networks and Possible Defencescited by 41
  14. 2021Trigger Hunting with a Topological Prior for Trojan Detectioncited by 39
  15. 2022Training with More Confidence: Mitigating Injected and Natural Backdoors During Trainingcited by 38
  16. 2023Backdoor Defense via Deconfounded Representation Learningcited by 38
  17. 2022Rethinking the Reverse-engineering of Trojan Triggerscited by 37
  18. 2021Dual-Key Multimodal Backdoors for Visual Question Answeringcited by 35
  19. 2024Test-Time Backdoor Attacks on Multimodal Large Language Modelscited by 35
  20. 2021Topological Detection of Trojaned Neural Networkscited by 34
  21. 2022Identifying a Training-Set Attack's Target Using Renormalized Influence Estimationcited by 32
  22. 2023Elijah: Eliminating Backdoors Injected in Diffusion Models via Distribution Shiftcited by 31
  23. 2022An Embarrassingly Simple Backdoor Attack on Self-supervised Learningcited by 30
  24. 2022Post-Training Detection of Backdoor Attacks for Two-Class and Multi-Attack Scenarioscited by 29
  25. 2023Label Poisoning is All You Needcited by 27
  26. 2023Distilling Cognitive Backdoor Patterns within an Imagecited by 20
  27. 2023Defending Against Patch-based Backdoor Attacks on Self-Supervised Learningcited by 19
  28. 2021Backdoor Attack through Frequency Domaincited by 18
  29. 2022A Survey of Neural Trojan Attacks and Defenses in Deep Learningcited by 17
  30. 2023UMD: Unsupervised Model Detection for X2X Backdoor Attackscited by 17
  31. 2022Universal Backdoor Attacks Detection via Adaptive Adversarial Probecited by 16
  32. 2021Enhancing Backdoor Attacks with Multi-Level MMD Regularizationcited by 15
  33. 2022Few-shot Backdoor Attacks via Neural Tangent Kernelscited by 15
  34. 2023BEAGLE: Forensics of Deep Learning Backdoor Attack for Better Defensecited by 15
  35. 2022Jigsaw Puzzle: Selective Backdoor Attack to Subvert Malware Classifierscited by 14
  36. 2023Defenses in Adversarial Machine Learning: A Surveycited by 14
  37. 2024Preference Poisoning Attacks on Reward Model Learningcited by 14
  38. 2022Backdoor Attacks on Vision Transformerscited by 13

50 most cited of 143 All 143 (JSON)