SanitAIs: Unsupervised Data Augmentation to Sanitize Trojaned Neural Networks
On the map: LLM safety and security → Backdoor and data poisoning attacks
Abstract (arXiv)
Self-supervised learning (SSL) methods have resulted in broad improvements to neural network performance by leveraging large, untapped collections of unlabeled data to learn generalized underlying structure. In this work, we harness unsupervised data augmentation (UDA), an SSL technique, to mitigate backdoor or Trojan attacks on deep neural networks. We show that UDA is more effective at removing trojans than current state-of-the-art methods for both feature space and point triggers, over a range of model architectures, trojans, and data quantities provided for trojan removal. These results demonstrate that UDA is both an effective and practical approach to mitigating the effects of backdoors on neural networks.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Placed in Backdoor and data poisoning attacks, one of 12 directions in LLM safety and security. Builds on 17 of its 28 references that are in the map. First posted on arXiv 9 Sep 2021.
Builds on
17 of 28 references.
References · 17
- 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
- 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
- 2016Wide Residual Networkscited by 3,057
- 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,875
- 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,324
- 2019Unsupervised Data Augmentation for Consistency Trainingcited by 1,057
- 2017BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chaincited by 1,052
- 2018Loss Surfaces, Mode Connectivity, and Fast Ensembling of DNNscited by 585
- 2018Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networkscited by 550
- 2021Neural Attention Distillation: Erasing Backdoor Triggers from Deep Neural Networkscited by 313
- 2020Comparing Rewinding and Fine-tuning in Neural Network Pruningcited by 238
- 2019A backdoor attack against LSTM-based text classification systemscited by 214
5 more
- 2017Neural Trojanscited by 151
- 2020Bridging Mode Connectivity in Loss Landscapes and Adversarial Robustnesscited by 130
- 2020The TrojAI Software Framework: An OpenSource tool for Embedding Trojans into Deep Learning Modelscited by 41
- 2020MaxUp: A Simple Way to Improve Generalization of Neural Network Trainingcited by 27
- 2018PoTrojan: powerful neural-level trojan designs in deep learning modelscited by 9
Built on by
Citing papers per year: 2026: 1.
1 paper.