mlsanityData to 7 Oct 2026

Dormant Neural Trojans

arXiv 2211.01808 ↗First posted 2 Nov 2022cs.CR cs.LGcited by 0

On the map: LLM safety and security → Backdoor and data poisoning attacks

Abstract (arXiv)

We present a novel methodology for neural network backdoor attacks. Unlike existing training-time attacks where the Trojaned network would respond to the Trojan trigger after training, our approach inserts a Trojan that will remain dormant until it is activated. The activation is realized through a specific perturbation to the network's weight parameters only known to the attacker. Our analysis and the experimental results demonstrate that dormant Trojaned networks can effectively evade detection by state-of-the-art backdoor detection methods.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Backdoor and data poisoning attacks, one of 12 directions in LLM safety and security. Builds on 42 of its 61 references that are in the map. First posted on arXiv 2 Nov 2022.

Builds on

42 of 61 references.

Closest ancestors · 2

Same direction, newest first

  1. 2021Rethinking the Backdoor Attacks' Triggers: A Frequency Perspective7 Apr
  2. 2021Targeted Attack against Deep Neural Networks via Flipping Limited Weight Bits21 Feb

Other references · 40

  1. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  2. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  3. 2016Densely Connected Convolutional Networkscited by 6,691
  4. 2013Intriguing properties of neural networkscited by 5,114
  5. 2017BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chaincited by 1,052
  6. 2017Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoningcited by 1,012
  7. 2018How To Backdoor Federated Learningcited by 745
  8. 2018Loss Surfaces, Mode Connectivity, and Fast Ensembling of DNNscited by 585
  9. 2018Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networkscited by 550
  10. 2015An Introduction to Convolutional Neural Networkscited by 510
  11. 2018Poison Frogs! Targeted Clean-Label Poisoning Attacks on Neural Networkscited by 500
  12. 2018Spectral Signatures in Backdoor Attackscited by 457
28 more
  1. 2018Detecting Backdoor Attacks on Deep Neural Networks by Activation Clusteringcited by 429
  2. 2020Backdoor Learning: A Surveycited by 412
  3. 2019Hidden Trigger Backdoor Attackscited by 359
  4. 2020Reflection Backdoor: A Natural Backdoor Attack on Deep Neural Networkscited by 341
  5. 2020Weight Poisoning Attacks on Pre-trained Modelscited by 318
  6. 2021Neural Attention Distillation: Erasing Backdoor Triggers from Deep Neural Networkscited by 313
  7. 2020Attack of the Tails: Yes, You Really Can Backdoor Federated Learningcited by 307
  8. 2019Label-Consistent Backdoor Attackscited by 285
  9. 2019Detecting AI Trojans Using Meta Neural Analysiscited by 192
  10. 2019Universal Litmus Patterns: Revealing Backdoor Attacks in CNNscited by 143
  11. 2020Bridging Mode Connectivity in Loss Landscapes and Adversarial Robustnesscited by 130
  12. 2021SPECTRE: Defending Against Backdoor Attacks Using Robust Statisticscited by 123
  13. 2019TABOR: A Highly Accurate Approach to Inspecting and Restoring Trojan Backdoors in AI Systemscited by 120
  14. 2019TBT: Targeted Neural Network Attack with Bit Trojancited by 112
  15. 2019Bit-Flip Attack: Crushing Neural Network with Progressive Bit Searchcited by 106
  16. 2021Backdoor Scanning for Deep Neural Networks through K-Arm Optimizationcited by 96
  17. 2019Defending Neural Backdoors via Generative Distribution Modelingcited by 82
  18. 2021Backdoor Attack in the Physical Worldcited by 82
  19. 2020DeepSweep: An Evaluation Framework for Mitigating DNN Backdoor Attacks using Data Augmentationcited by 75
  20. 2020Can Adversarial Weight Perturbations Inject Neural Backdoors?cited by 51
  21. 2019NeuronInspect: Detecting Backdoors in Neural Networks via Output Explanationscited by 49
  22. 2018Model-Reuse Attacks on Deep Learning Systemscited by 47
  23. 2020Backdoor Attacks against Transfer Learning with Pre-trained Deep Learning Modelscited by 43
  24. 2020TrojanZoo: Towards Unified, Holistic, and Practical Evaluation of Neural Backdoorscited by 30
  25. 2020Neural Network Laundering: Removing Black-Box Backdoor Watermarks from Deep Neural Networkscited by 24
  26. 2021Robust Backdoor Attacks against Deep Neural Networks in Real Physical Worldcited by 20
  27. 2021TOP: Backdoor Detection in Neural Networks via Transferability of Perturbationcited by 13
  28. 2021TAD: Trigger Approximation based Black-box Trojan Detection for AIcited by 8

Built on by

Not cited yet.