mlsanityData to 7 Oct 2026

Defense against Adversarial Attacks in NLP via Dirichlet Neighborhood Ensemble

arXiv 2006.11627 ↗First posted 20 Jun 2020cs.CL cs.LGcited by 26

Not on the map. Why?

Abstract (arXiv)

Despite neural networks have achieved prominent performance on many natural language processing (NLP) tasks, they are vulnerable to adversarial examples. In this paper, we propose Dirichlet Neighborhood Ensemble (DNE), a randomized smoothing method for training a robust model to defense substitution-based attacks. During training, DNE forms virtual sentences by sampling embedding vectors for each word in an input sentence from a convex hull spanned by the word and its synonyms, and it augments them with the training data. In such a way, the model is robust to adversarial attacks while maintaining the performance on the original clean data. DNE is agnostic to the network architectures and scales to large models for NLP applications. We demonstrate through extensive experimentation that our method consistently outperforms recently proposed defense methods by a significant margin across different network architectures and multiple data sets.

Key facts. Cited by 26 papers in the map (data through 7 Oct 2026). First posted on arXiv 20 Jun 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 9; 2022: 9; 2023: 2; 2024: 3; 2025: 1; 2026: 2.

26 papers.

Sort:

By area:

  1. 2021Towards Robustness Against Natural Language Word Substitutionscited by 68
  2. 2021Certified Robustness to Text Adversarial Attacks by Randomized [MASK]cited by 57
  3. 2021Searching for an Effective Defender: Benchmarking Defense against Adversarial Word Substitutioncited by 55
  4. 2024One Perturbation is Enough: On Generating Universal Adversarial Perturbations against Vision-Language Pre-training Modelscited by 39
  5. 2021Defending Against Backdoor Attacks in Natural Language Generationcited by 30
  6. 2021Grey-box Adversarial Attack And Defence For Sentiment Classificationcited by 28
  7. 2022"That Is a Suspicious Reaction!": Interpreting Logits Variation to Detect NLP Adversarial Attackscited by 24
  8. 2021Achieving Model Robustness through Discrete Adversarial Trainingcited by 23
  9. 2022A Survey of Adversarial Defences and Robustness in NLPcited by 22
  10. 2022Improving the Adversarial Robustness of NLP Models by Information Bottleneckcited by 17
  11. 2022Text Adversarial Purification as Defense against Adversarial Attackscited by 15
  12. 2022Defending Substitution-Based Profile Pollution Attacks on Sequential Recommenderscited by 11
14 more
  1. 2024Trustworthy Federated Learning: Privacy, Security, and Beyondcited by 11
  2. 2021Double Perturbation: On the Robustness of Robustness and Counterfactual Bias Evaluationcited by 9
  3. 2022Robust Natural Language Processing: Recent Advances, Challenges, and Future Directionscited by 9
  4. 2021Detecting Textual Adversarial Examples through Randomized Substitution and Votecited by 8
  5. 2023TextShield: Beyond Successfully Detecting Adversarial Sentences in Text Classificationcited by 8
  6. 2023Masked Language Model Based Textual Adversarial Example Detectioncited by 5
  7. 2024Extreme Miscalibration and the Illusion of Adversarial Robustnesscited by 4
  8. 2021Adversarial Machine Learning in Text Analysis and Generationcited by 2
  9. 2022Learning to Ignore Adversarial Attackscited by 2
  10. 2022ADDMU: Detection of Far-Boundary Adversarial Examples with Data and Model Uncertainty Estimationcited by 2
  11. 2025Bridging Robustness and Generalization Against Word Substitution Attacks in NLP via the Growth Bound Matrix Approachcited by 1
  12. 2022Probabilistic Categorical Adversarial Attack & Adversarial Trainingcited by 0
  13. 2026S-GBT: Smooth Growth Bound Tensor for Certified Robustness Against Word Substitution Attacks in NLPcited by 0
  14. 2026Evaluation of Adversarial Robustness in Arabic Language Modelscited by 0