• Home
  • Search
  • Lost In Translation: Generating Adversarial Examples Robust to Round-Trip Translation
  • Cite Icon2
  • https://doi.org/10.1109/icassp49357.2023.10094630Copy DOI Icon

Lost In Translation: Generating Adversarial Examples Robust to Round-Trip Translation

  • Jun 4, 2023
  • Neel Bhandari +1 more
Show More
  • Abstract
  • PDF
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Language Models today provide a high accuracy across a large number of\ndownstream tasks. However, they remain susceptible to adversarial attacks,\nparticularly against those where the adversarial examples maintain considerable\nsimilarity to the original text. Given the multilingual nature of text, the\neffectiveness of adversarial examples across translations and how machine\ntranslations can improve the robustness of adversarial examples remain largely\nunexplored. In this paper, we present a comprehensive study on the robustness\nof current text adversarial attacks to round-trip translation. We demonstrate\nthat 6 state-of-the-art text-based adversarial attacks do not maintain their\nefficacy after round-trip translation. Furthermore, we introduce an\nintervention-based solution to this problem, by integrating Machine Translation\ninto the process of adversarial example generation and demonstrating increased\nrobustness to round-trip translation. Our results indicate that finding\nadversarial examples robust to translation can help identify the insufficiency\nof language models that is common across languages, and motivate further\nresearch into multilingual adversarial attacks.\n

Loading PDF

Similar Papers
  • Research Article
  • Citations112

Adversarial examples: A survey of attacks and defenses in deep learning-enabled cybersecurity systems

  • Oct 20, 2023
  • Expert Systems with Applications
  • Mayra Macas +2
  • PDF
  • Research Article
  • Citations34

Adversarial Attack for SAR Target Recognition Based on UNet-Generative Adversarial Network

  • Oct 29, 2021
  • Remote Sensing
  • Chuan Du +1
  • Research Article
  • Citations11

Efficient Generation of Targeted and Transferable Adversarial Examples for Vision-Language Models via Diffusion Models

  • Jan 01, 2025
  • IEEE Transactions on Information Forensics and Security
  • Qi Guo +4
  • PDF
  • Research Article
  • Citations26

On the Effectiveness of Adversarial Training in Defending against Adversarial Example Attacks for Image Classification

  • Nov 14, 2020
  • Applied Sciences
  • Sanglee Park +1
  • Research Article
  • Citations12

A Word-Level Adversarial Attack Method Based on Sememes and an Improved Quantum-Behaved Particle Swarm Optimization.

  • Nov 01, 2024
  • IEEE Transactions on Neural Networks and Learning Systems
  • Qidong Chen +2
  • Research Article
  • Citations15

Feature-filter: Detecting adversarial examples by filtering out recessive features

  • May 20, 2022
  • Applied Soft Computing
  • Hui Liu +5
  • Conference Article
  • Citations1

ADVRET: An Adversarial Robustness Evaluating and Testing Platform for Deep Learning Models

  • Dec 01, 2021
  • Fei Ren +4
  • Research Article
  • Citations7

SAR-PATT: A Physical Adversarial Attack for SAR Image Automatic Target Recognition

  • Dec 25, 2024
  • Remote Sensing
  • Binyan Luo +6
  • PDF
  • Research Article
  • Citations7

Adversarial attacks on cooperative multi-agent deep reinforcement learning: a dynamic group-based adversarial example transferability method

  • Jul 03, 2023
  • Complex & Intelligent Systems
  • Lixia Zan +2
  • Research Article
  • Citations3

Explanation-Guided Adversarial Example Attacks

  • Mar 26, 2024
  • Big Data Research
  • Anli Yan +4
  • PDF
  • Conference Article
  • Citations5

Second-Order NLP Adversarial Examples

  • Jan 01, 2020
  • John Morris
  • Conference Article
  • Citations3

The Best Defense is Attack: Repairing Semantics in Textual Adversarial Examples

  • Jan 01, 2024
  • Heng Yang +1
  • PDF
  • Research Article
  • Citations17

Diversity Adversarial Training against Adversarial Attack on Deep Neural Networks

  • Mar 06, 2021
  • Symmetry
  • Hyun Kwon +1
  • Conference Article

Hidden Conditional Adversarial Attacks

  • Oct 16, 2022
  • Junyoung Byun +3
  • PDF
  • Research Article
  • Citations15

Detection of adversarial attacks based on differences in image entropy

  • Aug 17, 2023
  • International Journal of Information Security
  • Gwonsang Ryu +1
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.