Robust Neural Machine Translation with Doubly Adversarial Inputs

@inproceedings{Cheng2019RobustNM,
  title={Robust Neural Machine Translation with Doubly Adversarial Inputs},
  author={Yong Cheng and Lu Jiang and Wolfgang Macherey},
  booktitle={ACL},
  year={2019}
}
  • Yong Cheng, Lu Jiang, Wolfgang Macherey
  • Published in ACL 2019
  • Computer Science
  • Neural machine translation (NMT) often suffers from the vulnerability to noisy perturbations in the input. [...] Key MethodFor the generation of adversarial inputs, we propose a gradient-based method to craft adversarial examples informed by the translation loss over the clean inputs.Experimental results on Chinese-English and English-German translation tasks demonstrate that our approach achieves significant improvements ($2.8$ and $1.6$ BLEU points) over Transformer on standard clean benchmarks as well as…Expand Abstract
    62 Citations

    Tables and Topics from this paper

    Robust Unsupervised Neural Machine Translation with Adversarial Training
    • 2
    • Highly Influenced
    A Reinforced Generation of Adversarial Samples for Neural Machine Translation
    • 5
    • PDF
    AdvAug: Robust Adversarial Augmentation for Neural Machine Translation
    • 8
    • PDF
    Robust Unsupervised Neural Machine Translation with Adversarial Denoising Training
    • 1
    • PDF
    Adversarial Subword Regularization for Robust Neural Machine Translation
    • Highly Influenced
    • PDF
    FreeLB: Enhanced Adversarial Training for Language Understanding
    • 30
    • PDF
    FreeLB: Enhanced Adversarial Training for Natural Language Understanding
    • 58
    Improving Neural Machine Translation Robustness via Data Augmentation: Beyond Back Translation
    • 5
    • PDF

    References

    SHOWING 1-10 OF 36 REFERENCES
    Towards Robust Neural Machine Translation
    • 84
    • PDF
    On Adversarial Examples for Character-Level Neural Machine Translation
    • 69
    • Highly Influential
    • PDF
    Explaining and Harnessing Adversarial Examples
    • 6,521
    • PDF
    Generating Natural Language Adversarial Examples
    • 255
    • PDF
    Generating Natural Adversarial Examples
    • 268
    • PDF
    Synthetic and Natural Noise Both Break Neural Machine Translation
    • 288
    • PDF
    Adversarial Training Methods for Semi-Supervised Text Classification
    • 352
    • Highly Influential
    • PDF
    Training on Synthetic Noise Improves Robustness to Natural Noise in Machine Translation
    • 34
    • PDF
    Understanding Back-Translation at Scale
    • 376
    • PDF
    Toward Robust Neural Machine Translation for Noisy Input Sequences
    • 41
    • PDF