mlsanityData to 14 Oct 2026

NMT-Obfuscator Attack: Ignore a sentence in translation with only one word

arXiv 2411.12473 ↗First posted 19 Nov 2024cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

Neural Machine Translation systems are used in diverse applications due to their impressive performance. However, recent studies have shown that these systems are vulnerable to carefully crafted small perturbations to their inputs, known as adversarial attacks. In this paper, we propose a new type of adversarial attack against NMT models. In this attack, we find a word to be added between two sentences such that the second sentence is ignored and not translated by the NMT model. The word added between the two sentences is such that the whole adversarial text is natural in the source language. This type of attack can be harmful in practical scenarios since the attacker can hide malicious information in the automatic translation made by the target NMT model. Our experiments show that different NMT models and translation tasks are vulnerable to this type of attack. Our attack can successfully force the NMT models to ignore the second part of the input in the translation for more than 50% of all cases while being able to maintain low perplexity for the whole input.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 23 of its 31 references that are in the map. First posted on arXiv 19 Nov 2024.

Builds on

23 of 31 references.

References · 23

  1. 2017Attention Is All You Needcited by 60,361
  2. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,512
  3. 2014Neural Machine Translation by Jointly Learning to Align and Translatecited by 4,548
  4. 2018A Call for Clarity in Reporting BLEU Scorescited by 1,799
  5. 2019Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailmentcited by 671
  6. 2018Generating Natural Language Adversarial Examplescited by 424
  7. 2018Black-box Generation of Adversarial Text Sequences to Evade Deep Learning Classifierscited by 345
  8. 2020Multilingual Translation with Extensible Multilingual Pretraining and Finetuningcited by 287
  9. 2021Gradient-based Adversarial Attacks against Text Transformerscited by 234
  10. 2018Marian: Fast Neural Machine Translation in C++cited by 203
  11. 2019Combating Adversarial Misspellings with Robust Word Recognitioncited by 163
  12. 2019Robust Neural Machine Translation with Doubly Adversarial Inputscited by 106
11 more
  1. 2018Seq2Sick: Evaluating the Robustness of Sequence-to-Sequence Models with Adversarial Examplescited by 94
  2. 2018On Adversarial Examples for Character-Level Neural Machine Translationcited by 85
  3. 2020Imitation Attacks and Defenses for Black-box Machine Translation Systemscited by 81
  4. 2019On Evaluation of Adversarial Perturbations for Sequence-to-Sequence Modelscited by 56
  5. 2020Generating Natural Language Attacks in a Hard Label Black Box Settingcited by 50
  6. 2020Evaluating Robustness to Input Perturbations for Neural Machine Translationcited by 36
  7. 2021Bridge the Gap Between CV and NLP! A Gradient-based Textual Adversarial Attack Frameworkcited by 28
  8. 2023TransFool: An Adversarial Attack against Neural Machine Translation Modelscited by 12
  9. 2023Targeted Adversarial Attacks against Neural Machine Translationcited by 10
  10. 2022Block-Sparse Adversarial Attack to Fool Transformer-Based Text Classifierscited by 9
  11. 2023A Classification-Guided Approach for Adversarial Attacks against Neural Machine Translationcited by 4

Built on by

Not cited yet.