mlsanityData to 7 Oct 2026

Natural Language Adversarial Defense through Synonym Encoding

arXiv 1909.06723 ↗First posted 15 Sep 2019cs.CL cs.LGcited by 39

Not on the map. Why?

Abstract (arXiv)

In the area of natural language processing, deep learning models are recently known to be vulnerable to various types of adversarial perturbations, but relatively few works are done on the defense side. Especially, there exists few effective defense method against the successful synonym substitution based attacks that preserve the syntactic structure and semantic information of the original text while fooling the deep learning models. We contribute in this direction and propose a novel adversarial defense method called Synonym Encoding Method (SEM). Specifically, SEM inserts an encoder before the input layer of the target model to map each cluster of synonyms to a unique encoding and trains the model to eliminate possible adversarial perturbations without modifying the network architecture or adding extra data. Extensive experiments demonstrate that SEM can effectively defend the current synonym substitution based attacks and block the transferability of adversarial examples. SEM is also easy and efficient to scale to large models and big datasets.

Key facts. Cited by 39 papers in the map (data through 7 Oct 2026). First posted on arXiv 15 Sep 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 4; 2022: 6; 2023: 12; 2024: 8; 2025: 6; 2026: 3.

39 papers.

Sort:

By area:

  1. 2023SmoothLLM: Defending Large Language Models Against Jailbreaking Attackscited by 434
  2. 2022Why Should Adversarial Perturbations be Imperceptible? Rethink the Research Paradigm in Adversarial NLPcited by 99
  3. 2024Defending Against Unforeseen Failure Modes with Latent Adversarial Trainingcited by 82
  4. 2023Structure Invariant Transformation for better Adversarial Transferabilitycited by 61
  5. 2021Triangle Attack: A Query-efficient Decision-based Adversarial Attackcited by 48
  6. 2023Text-CRS: A Generalized Certified Robustness Framework against Textual Adversarial Attackscited by 32
  7. 2023The Trickle-down Impact of Reward (In-)consistency on RLHFcited by 29
  8. 2024On the Duality Between Sharpness-Aware Minimization and Adversarial Trainingcited by 24
  9. 2022A Survey of Adversarial Defences and Robustness in NLPcited by 22
  10. 2021Token-Modification Adversarial Attacks for Natural Language Processing: A Surveycited by 12
  11. 2022Robust Textual Embedding against Word-level Adversarial Attackscited by 12
  12. 2022Adversarial Training for Improving Model Robustness? Look at Both Prediction and Interpretationcited by 12
27 more
  1. 2022TextHacker: Learning based Hybrid Local Search Algorithm for Text Hard-label Adversarial Attackcited by 10
  2. 2024Spiking Convolutional Neural Networks for Text Classificationcited by 10
  3. 2023TextDefense: Adversarial Text Detection based on Word Importance Entropycited by 9
  4. 2023Model-tuning Via Prompts Makes NLP Models Adversarially Robustcited by 9
  5. 2021Detecting Textual Adversarial Examples through Randomized Substitution and Votecited by 8
  6. 2023TextShield: Beyond Successfully Detecting Adversarial Sentences in Text Classificationcited by 8
  7. 2021How Vulnerable Are Automatic Fake News Detection Methods to Adversarial Attacks?cited by 7
  8. 2024Fast Adversarial Training against Textual Adversarial Attackscited by 7
  9. 2023Masked Language Model Based Textual Adversarial Example Detectioncited by 5
  10. 2022Using Random Perturbations to Mitigate Adversarial Attacks on Sentiment Analysis Modelscited by 4
  11. 2025Exploring Gradient-Guided Masked Language Model to Detect Textual Adversarial Attackscited by 4
  12. 2023The Best Defense is Attack: Repairing Semantics in Textual Adversarial Examplescited by 3
  13. 2023From Adversarial Arms Race to Model-centric Evaluation: Motivating a Unified Automatic Robustness Evaluation Frameworkcited by 3
  14. 2024SemRoDe: Macro Adversarial Training to Learn Representations That are Robust to Word-Level Attackscited by 3
  15. 2025Guided Perturbation Sensitivity (GPS): Detecting Adversarial Text via Embedding Stability and Word Importancecited by 2
  16. 2023Alert-ME: An Explainability-Driven Defense Against Adversarial Examples in Transformer-Based Text Classificationcited by 1
  17. 2025Bridging Robustness and Generalization Against Word Substitution Attacks in NLP via the Growth Bound Matrix Approachcited by 1
  18. 2025Bounded PCTL Model Checking of Large Language Model Outputscited by 1
  19. 2023VoteTRANS: Detecting Adversarial Text without Training by Voting on Hard Labels of Transformationscited by 0
  20. 2024Single Word Change is All You Need: Using LLMs to Create Synthetic Training Examples for Text Classifierscited by 0
  21. 2024READ: Improving Relation Extraction from an ADversarial Perspectivecited by 0
  22. 2024GenFighter: A Generative and Evolutive Textual Attack Removalcited by 0
  23. 2025DINA: A Dual Defense Framework Against Internal Noise and External Attacks in Natural Language Processingcited by 0
  24. 2025Overcoming Black-box Attack Inefficiency with Hybrid and Dynamic Select Algorithmscited by 0
  25. 2026S-GBT: Smooth Growth Bound Tensor for Certified Robustness Against Word Substitution Attacks in NLPcited by 0
  26. 2026Build it, Break it, Repeat: Benchmarking and improving LLM-manipulated disinformation detection in social media postscited by 0
  27. 2026BRANCH: Bypassing Multi-Scanner AI Guardrailscited by 0