mlsanityData to 14 Oct 2026

VoteTRANS: Detecting Adversarial Text without Training by Voting on Hard Labels of Transformations

arXiv 2306.01273 ↗First posted 2 Jun 2023cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

Adversarial attacks reveal serious flaws in deep learning models. More dangerously, these attacks preserve the original meaning and escape human recognition. Existing methods for detecting these attacks need to be trained using original/adversarial data. In this paper, we propose detection without training by voting on hard labels from predictions of transformations, namely, VoteTRANS. Specifically, VoteTRANS detects adversarial text by comparing the hard labels of input text and its transformation. The evaluation demonstrates that VoteTRANS effectively detects adversarial text across various state-of-the-art attacks, models, and datasets.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 21 of its 29 references that are in the map. First posted on arXiv 2 Jun 2023.

Builds on

21 of 29 references.

References · 21

  1. 2020Beyond Accuracy: Behavioral Testing of NLP models with CheckListcited by 1,006
  2. 2019Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailmentcited by 671
  3. 2020BERT-ATTACK: Adversarial Attack Against BERT Using BERTcited by 491
  4. 2018Generating Natural Language Adversarial Examplescited by 424
  5. 2018TextBugger: Generating Adversarial Text Against Real-world Applicationscited by 381
  6. 2018Black-box Generation of Adversarial Text Sequences to Evade Deep Learning Classifierscited by 345
  7. 2020BAE: BERT-based Adversarial Examples for Text Classificationcited by 316
  8. 2018Adversarial Example Generation with Syntactically Controlled Paraphrase Networkscited by 286
  9. 2019Certified Robustness to Adversarial Word Substitutionscited by 168
  10. 2019Combating Adversarial Misspellings with Robust Word Recognitioncited by 163
  11. 2018Pathologies of Neural Models Make Interpretations Difficultcited by 140
  12. 2020Contextualized Perturbation for Textual Adversarial Attackcited by 136
9 more
  1. 2021Towards Improving Adversarial Training of NLP Modelscited by 79
  2. 2019Learning to Discriminate Perturbations for Blocking Adversarial Attacks in Text Classificationcited by 57
  3. 2020Frequency-Guided Word Substitutions for Detecting Textual Adversarial Examplescited by 54
  4. 2019Natural Language Adversarial Defense through Synonym Encodingcited by 40
  5. 2022Detection of Word Adversarial Examples in Text Classification: Benchmark and Baseline via Robust Density Estimationcited by 32
  6. 2022"That Is a Suspicious Reaction!": Interpreting Logits Variation to Detect NLP Adversarial Attackscited by 24
  7. 2022Residue-Based Natural Language Adversarial Attack Detectioncited by 8
  8. 2022Distinguishing Non-natural from Natural Adversarial Samples for More Robust Pre-trained Language Modelcited by 3
  9. 2022Input-specific Attention Subnetworks for Adversarial Detectioncited by 2

Built on by

Not cited yet.