mlsanityData to 7 Oct 2026

TrojanDec: Data-free Detection of Trojan Inputs in Self-supervised Learning

arXiv 2501.04108 ↗First posted 7 Jan 2025cs.CR cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

An image encoder pre-trained by self-supervised learning can be used as a general-purpose feature extractor to build downstream classifiers for various downstream tasks. However, many studies showed that an attacker can embed a trojan into an encoder such that multiple downstream classifiers built based on the trojaned encoder simultaneously inherit the trojan behavior. In this work, we propose TrojanDec, the first data-free method to identify and recover a test input embedded with a trigger. Given a (trojaned or clean) encoder and a test input, TrojanDec first predicts whether the test input is trojaned. If not, the test input is processed in a normal way to maintain the utility. Otherwise, the test input will be further restored to remove the trigger. Our extensive evaluation shows that TrojanDec can effectively identify the trojan (if any) from a given test input and recover it under state-of-the-art trojan attacks. We further demonstrate by experiments that our TrojanDec outperforms the state-of-the-art defenses.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 32 of its 53 references that are in the map. First posted on arXiv 7 Jan 2025.

Builds on

32 of 53 references.

References · 32

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  3. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  4. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,482
  5. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  6. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,712
  7. 2017Improved Training of Wasserstein GANscited by 2,223
  8. 2017EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classificationcited by 1,488
  9. 2018Learning deep representations by mutual information estimation and maximizationcited by 1,265
  10. 2017BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chaincited by 1,052
  11. 2017Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoningcited by 1,012
  12. 2022Zero-Shot Image Restoration Using Denoising Diffusion Null-Space Modelcited by 574
20 more
  1. 2018Spectral Signatures in Backdoor Attackscited by 457
  2. 2019STRIP: A Defence Against Trojan Attacks on Deep Neural Networkscited by 437
  3. 2018Detecting Backdoor Attacks on Deep Neural Networks by Activation Clusteringcited by 429
  4. 2020Reflection Backdoor: A Natural Backdoor Attack on Deep Neural Networkscited by 341
  5. 2020Invisible Backdoor Attack with Sample-Specific Triggerscited by 338
  6. 2019Detecting AI Trojans Using Meta Neural Analysiscited by 192
  7. 2021Adversarial Unlearning of Backdoors via Implicit Hypergradientcited by 160
  8. 2021Rethinking the Backdoor Attacks' Triggers: A Frequency Perspectivecited by 155
  9. 2022Backdoor Defense via Decoupling the Training Processcited by 149
  10. 2022The "Beatrix'' Resurrections: Robust Backdoor Detection via Gram Matricescited by 62
  11. 2020Certified Robustness of Nearest Neighbors against Data Poisoning and Backdoor Attackscited by 58
  12. 2022MM-BD: Post-Training Detection of Backdoor Attacks with Arbitrary Backdoor Pattern Types Using a Maximum Margin Statisticcited by 53
  13. 2023Detecting Backdoors in Pre-trained Encoderscited by 53
  14. 2022PoisonedEncoder: Poisoning the Unlabeled Pre-training Data in Contrastive Learningcited by 33
  15. 2023Defending Pre-trained Language Models as Few-shot Learners against Backdoor Attackscited by 32
  16. 2022CorruptEncoder: Data Poisoning based Backdoor Attacks to Contrastive Learningcited by 28
  17. 2022Architectural Backdoors in Neural Networkscited by 24
  18. 2024Backdoor Contrastive Learning via Bi-level Trigger Optimizationcited by 14
  19. 2023On the Difficulty of Defending Contrastive Learning against Backdoor Attackscited by 8
  20. 2024Defending Deep Regression Models against Backdoor Attackscited by 3

Built on by

Not cited yet.