Toward Robust Neural Machine Translation for Noisy Input Sequences

By Matthias Sperber, Jan Niehues, A. Waibel. 2017

81 citations recorded by the source corpus.

Subfields: Safety, Robustness & Interpretability, Machine Translation

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Safety, Robustness & Interpretability

1. Lu Jiang
2. Wolfgang Macherey
3. Matthias Paulik
4. Yong Cheng
5. Jacob Eisenstein
6. Graham Neubig
7. Marcis Pinnis
8. Matthias Sperber

In Machine Translation

1. Matthias Paulik
2. Yong Cheng
3. Graham Neubig
4. Matthias Sperber
5. Wolfgang Macherey
6. Lu Jiang

Citations in the corpus

• Europarl-ST: A Multilingual Corpus for Speech Translation of Parliamentary Debates (2019, methodology): “This adaptation can be carried out by training MT systems on real ASR output as source input [27] or on simulated ASR output by applying noising techniques to the source side [1].”
• Speech Translation and the End-to-End Promise: Taking Stock of Where We Are (2020, methodology): “One way of achieving this is to train robust translation models by introducing synthetic ASR errors into the source side of MT corpora (Peitz et al., 2012; Tsvetkov et al., 2014; Ruiz et al., 2015; Sperber et al., 2017b; Cheng et al., 2018, 2019).”
• Data Augmentation for Pipeline-Based Speech Translation (2020, background): “The adverse effects of error propagation in pipeline-based speech translation systems have been studied [7] and addressed before [4,8,9,10].”
• Improving Robustness of Machine Translation with Synthetic Noise (2019, background): “Experiments that specifically handle noise include text normalization approaches (Baldwin et al., 2015) and (most relevant to our work) the artificial induction of noise in otherwise clean data (Sperber et al., 2017; Belinkov and Bisk, 2017).”
• AdvAug: Robust Adversarial Augmentation for Neural Machine Translation (2020, background): “, 2019), and (2) discrete noise which adds, deletes, and/or replaces characters or words in the observed sentences (Belinkov and Bisk, 2018; Sperber et al., 2017; Ebrahimi et al., 2018; Michel et al., 2019; Cheng et al., 2019; Karpukhin et al., 2019).”