mlsanityData to 7 Oct 2026

A General Framework for Defending Against Backdoor Attacks via Influence Graph

arXiv 2111.14309 ↗First posted 29 Nov 2021cs.LG cs.AI cs.CLcited by 1

Not on the map. Why?

Abstract (arXiv)

In this work, we propose a new and general framework to defend against backdoor attacks, inspired by the fact that attack triggers usually follow a textscspecific type of attacking pattern, and therefore, poisoned training examples have greater impacts on each other during training. We introduce the notion of the it influence graph, which consists of nodes and edges respectively representative of individual training points and associated pair-wise influences. The influence between a pair of training points represents the impact of removing one training point on the prediction of another, approximated by the influence function citepkoh2017understanding. Malicious training points are extracted by finding the maximum average sub-graph subject to a particular size. Extensive experiments on computer vision and natural language processing tasks demonstrate the effectiveness and generality of the proposed framework.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 49 of its 67 references that are in the map. First posted on arXiv 29 Nov 2021.

Builds on

49 of 67 references.

References · 49

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  3. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  4. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  5. 2015Character-level Convolutional Networks for Text Classificationcited by 2,293
  6. 2017Understanding Black-box Predictions via Influence Functionscited by 1,714
  7. 2017BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chaincited by 1,052
  8. 2017Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoningcited by 1,012
  9. 2018How To Backdoor Federated Learningcited by 745
  10. 2018Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networkscited by 550
  11. 2018Spectral Signatures in Backdoor Attackscited by 457
  12. 2019STRIP: A Defence Against Trojan Attacks on Deep Neural Networkscited by 437
37 more
  1. 2018Detecting Backdoor Attacks on Deep Neural Networks by Activation Clusteringcited by 429
  2. 2019Hidden Trigger Backdoor Attackscited by 359
  3. 2020Reflection Backdoor: A Natural Backdoor Attack on Deep Neural Networkscited by 341
  4. 2020Weight Poisoning Attacks on Pre-trained Modelscited by 318
  5. 2020Input-Aware Dynamic Backdoor Attackcited by 304
  6. 2019Label-Consistent Backdoor Attackscited by 285
  7. 2020ONION: A Simple and Effective Defense Against Textual Backdoor Attackscited by 231
  8. 2019A backdoor attack against LSTM-based text classification systemscited by 214
  9. 2021Hidden Killer: Invisible Textual Backdoor Attacks with Syntactic Triggercited by 203
  10. 2020Clean-Label Backdoor Attacks on Video Recognition Modelscited by 162
  11. 2018SentiNet: Detecting Localized Universal Attacks Against Deep Learning Systemscited by 148
  12. 2019Universal Litmus Patterns: Revealing Backdoor Attacks in CNNscited by 143
  13. 2018Backdoor Embedding in Convolutional Neural Network Models via Invisible Perturbationcited by 131
  14. 2021SPECTRE: Defending Against Backdoor Attacks Using Robust Statisticscited by 123
  15. 2021BadEncoder: Backdoor Attacks to Pre-trained Encoders in Self-Supervised Learningcited by 120
  16. 2020RAB: Provable Robustness Against Backdoor Attackscited by 115
  17. 2021Be Careful about Poisoned Word Embeddings: Exploring the Vulnerability of the Embedding Layers in NLP Modelscited by 96
  18. 2019Robust Anomaly Detection and Backdoor Attack Detection Via Differential Privacycited by 93
  19. 2021Turn the Combination Lock: Learnable Textual Backdoor Attacks via Word Substitutioncited by 85
  20. 2019Defending Neural Backdoors via Generative Distribution Modelingcited by 82
  21. 2020On the Effectiveness of Mitigating Data Poisoning Attacks with Gradient Shapingcited by 82
  22. 2020Mitigating backdoor attacks in LSTM-based Text Classification Systems by Backdoor Keyword Identificationcited by 80
  23. 2020Trojaning Language Models for Fun and Profitcited by 80
  24. 2021BadPre: Task-agnostic Backdoor Attacks to Pre-trained NLP Foundation Modelscited by 76
  25. 2020On Certifying Robustness against Backdoor Attacks via Randomized Smoothingcited by 74
  26. 2020Backdoor Attack against Speaker Verificationcited by 62
  27. 2021Triggerless Backdoor Attack for NLP Tasks with Clean Labelscited by 61
  28. 2021Poison Ink: Robust and Invisible Backdoor Attackcited by 50
  29. 2020Backdoor Attacks against Transfer Learning with Pre-trained Deep Learning Modelscited by 43
  30. 2019Quantum Entropy Scoring for Fast Robust Mean Estimation and Improved Outlier Detectioncited by 41
  31. 2020One-pixel Signature: Characterizing CNN Models for Backdoor Detectioncited by 34
  32. 2021Defending Against Backdoor Attacks in Natural Language Generationcited by 30
  33. 2020Backdooring and Poisoning Neural Networks with Image-Scaling Attackscited by 28
  34. 2020ConFoc: Content-Focus Protection Against Trojan Attacks on Neural Networkscited by 28
  35. 2020BAAAN: Backdoor Attacks Against Autoencoder and GAN-Based Machine Learning Modelscited by 23
  36. 2020Pair the Dots: Jointly Examining Training History and Test Stimuli for Model Interpretabilitycited by 7
  37. 2021Spinning Sequence-to-Sequence Models with Meta-Backdoorscited by 5

Built on by

Cited by year

Citing papers per year: 2023: 1.

1 paper.

  1. 2023IMBERT: Making BERT Immune to Insertion-based Backdoor Attackscited by 12