mlsanityData to 14 Oct 2026

Residue-Based Natural Language Adversarial Attack Detection

arXiv 2204.10192 ↗First posted 17 Apr 2022cs.CL cs.AIcited by 8

Not on the map. Why?

Abstract (arXiv)

Deep learning based systems are susceptible to adversarial attacks, where a small, imperceptible change at the input alters the model prediction. However, to date the majority of the approaches to detect these attacks have been designed for image processing systems. Many popular image adversarial detection approaches are able to identify adversarial examples from embedding feature spaces, whilst in the NLP domain existing state of the art detection approaches solely focus on input text features, without consideration of model embedding spaces. This work examines what differences result when porting these image designed strategies to Natural Language Processing (NLP) tasks - these detectors are found to not port over well. This is expected as NLP systems have a very different form of input: discrete and sequential in nature, rather than the continuous and fixed size inputs for images. As an equivalent model-focused NLP detection approach, this work proposes a simple sentence-embedding "residue" based detector to identify adversarial examples. On many tasks, it out-performs ported image domain detectors and recent state of the art NLP specific detectors.

Key facts. Cited by 8 papers in the map (data through 14 Oct 2026). Builds on 40 of its 66 references that are in the map. First posted on arXiv 17 Apr 2022.

Builds on

40 of 66 references.

References · 40

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  3. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  4. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  5. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  6. 2013Intriguing properties of neural networkscited by 5,155
  7. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,448
  8. 2015Character-level Convolutional Networks for Text Classificationcited by 2,330
  9. 2018A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attackscited by 1,485
  10. 2016Adversarial Machine Learning at Scalecited by 936
  11. 2016Universal adversarial perturbationscited by 784
  12. 2017Adversarial Examples for Evaluating Reading Comprehension Systemscited by 571
28 more
  1. 2018Universal Sentence Encodercited by 489
  2. 2017Wild Patterns: Ten Years After the Rise of Adversarial Machine Learningcited by 414
  3. 2018TextBugger: Generating Adversarial Text Against Real-world Applicationscited by 381
  4. 2018Adversarial Example Generation with Syntactically Controlled Paraphrase Networkscited by 286
  5. 2018Characterizing Adversarial Subspaces Using Local Intrinsic Dimensionalitycited by 282
  6. 2017Detecting Adversarial Samples from Artifactscited by 245
  7. 2017Generating Natural Adversarial Examplescited by 183
  8. 2018Understanding Measures of Uncertainty for Adversarial Example Detectioncited by 177
  9. 2016Crafting Adversarial Input Sequences for Recurrent Neural Networkscited by 103
  10. 2018Seq2Sick: Evaluating the Robustness of Sequence-to-Sequence Models with Adversarial Examplescited by 94
  11. 2021Towards Improving Adversarial Training of NLP Modelscited by 79
  12. 2016Adversarial Examples Detection in Deep Networks with Convolutional Filter Statisticscited by 77
  13. 2019Detecting Adversarial Samples Using Influence Functions and Nearest Neighborscited by 72
  14. 2019Learning to Discriminate Perturbations for Blocking Adversarial Attacks in Text Classificationcited by 57
  15. 2017Towards Crafting Text Adversarial Samplescited by 55
  16. 2020Frequency-Guided Word Substitutions for Detecting Textual Adversarial Examplescited by 54
  17. 2018Adversarially Regularising Neural NLI Models to Integrate Logical Background Knowledgecited by 46
  18. 2018Robust Machine Comprehension Models via Adversarial Trainingcited by 44
  19. 2016Adversarial Perturbations Against Deep Neural Networks for Malware Classificationcited by 42
  20. 2018Adversarial Deep Learning for Robust Detection of Binary Encoded Malwarecited by 32
  21. 2018Adversarial Over-Sensitivity and Over-Stability Strategies for Dialogue Modelscited by 26
  22. 2020Adversarial Attack and Defense of Structured Prediction Modelscited by 20
  23. 2018Comparing Attention-based Convolutional and Recurrent Neural Networks: Success and Limitations in Machine Reading Comprehensioncited by 19
  24. 2018Identify Susceptible Locations in Medical Records via Adversarial Attacks on Deep Predictive Modelscited by 15
  25. 2017Grammatical Error Correction with Neural Reinforcement Learningcited by 7
  26. 2019Learning To Characterize Adversarial Subspacescited by 5
  27. 2017Generic Black-Box End-to-End Attack Against State of the Art API Call Based Malware Classifierscited by 4
  28. 2018Adversarial Reprogramming of Text Classification Neural Networkscited by 3

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 5; 2024: 1; 2025: 1.

8 papers.

  1. 2023Identifying Adversarially Attackable and Robust Samplescited by 5
  2. 2023Whispers of Doubt Amidst Echoes of Triumph in NLP Robustnesscited by 5
  3. 2024Self-Supervised Representation Learning for Adversarial Attack Detectioncited by 4
  4. 2023Sample Attackability in Natural Language Adversarial Attackscited by 2
  5. 2025NLP Security and Ethics, in the Wildcited by 2
  6. 2023Sentiment Perception Adversarial Attacks on Neural Machine Translation Systemscited by 1
  7. 2022Gender Bias and Universal Substitution Adversarial Attacks on Grammatical Error Correction Systems for Automated Assessmentcited by 0
  8. 2023VoteTRANS: Detecting Adversarial Text without Training by Voting on Hard Labels of Transformationscited by 0