mlsanityData to 14 Oct 2026

Step by Step Loss Goes Very Far: Multi-Step Quantization for Adversarial Text Attacks

arXiv 2302.05120 ↗First posted 10 Feb 2023cs.CLcited by 3

Not on the map. Why?

Abstract (arXiv)

We propose a novel gradient-based attack against transformer-based language models that searches for an adversarial example in a continuous space of token probabilities. Our algorithm mitigates the gap between adversarial loss for continuous and discrete text representations by performing multi-step quantization in a quantization-compensation loop. Experiments show that our method significantly outperforms other approaches on various natural language processing (NLP) tasks.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 23 of its 34 references that are in the map. First posted on arXiv 10 Feb 2023.

Builds on

23 of 34 references.

References · 23

  1. 2017Attention Is All You Needcited by 60,361
  2. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  3. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  4. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,512
  5. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,998
  6. 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,739
  7. 2017A Broad-Coverage Challenge Corpus for Sentence Understanding through Inferencecited by 2,682
  8. 2015Character-level Convolutional Networks for Text Classificationcited by 2,330
  9. 2019Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailmentcited by 671
  10. 2017ZOO: Zeroth Order Optimization based Black-box Attacks to Deep Neural Networks without Training Substitute Modelscited by 666
  11. 2020BERT-ATTACK: Adversarial Attack Against BERT Using BERTcited by 491
  12. 2018Universal Sentence Encodercited by 489
11 more
  1. 2020BAE: BERT-based Adversarial Examples for Text Classificationcited by 316
  2. 2021Recent Advances in Adversarial Training for Adversarial Robustnesscited by 279
  3. 2021Gradient-based Adversarial Attacks against Text Transformerscited by 234
  4. 2020A Primer on Zeroth-Order Optimization in Signal Processing and Machine Learningcited by 162
  5. 2019Robust Neural Machine Translation with Doubly Adversarial Inputscited by 106
  6. 2020Reevaluating Adversarial Examples in Natural Languagecited by 85
  7. 2019ZO-AdaMM: Zeroth-Order Adaptive Momentum Method for Black-Box Optimizationcited by 76
  8. 2017Stochastic Zeroth-order Optimization in High Dimensionscited by 63
  9. 2021Bridge the Gap Between CV and NLP! A Gradient-based Textual Adversarial Attack Frameworkcited by 28
  10. 2020Searching for a Search Method: Benchmarking Search Algorithms for Generating NLP Adversarial Examplescited by 23
  11. 2021Don't Search for a Search Method -- Simple Heuristics Suffice for Adversarial Text Attackscited by 5

Built on by

Cited by year

Citing papers per year: 2023: 1; 2024: 2.

3 papers.

  1. 2024Breaking Agents: Compromising Autonomous LLM Agents Through Malfunction Amplificationcited by 81
  2. 2023Privacy in Large Language Models: Attacks, Defenses and Future Directionscited by 51
  3. 2024Pointing out the Shortcomings of Relation Extraction Models with Semantically Motivated Adversarialscited by 1

Similar, not cited

Close in wording, no citation either way

  1. 2024The Impact of Quantization on the Robustness of Transformer-based Text Classifiers
  2. 2026Adversarial Contrastive Learning for LLM Quantization Attacks
  3. 2024Exploiting LLM Quantization
  4. 2024Adversarial Attacks on Large Language Models Using Regularized Relaxation
  5. 2024MaskPure: Improving Defense Against Text Adversaries with Stochastic Purification
  6. 2024A Constraint-Enforcing Reward for Adversarial Attacks on Text Classifiers
  7. 2023Baseline Defenses for Adversarial Attacks Against Aligned Language Models
  8. 2021Qu-ANTI-zation: Exploiting Quantization Artifacts for Achieving Adversarial Outcomes