Neural Machine Translation of Text from Non-Native Speakers

Antonios Anastasopoulos, Alison Lui, Toan Nguyen, David Chiang
8/19/2018
cs.CL

No confident code implementation found for this paper.

Abstract

Neural Machine Translation (NMT) systems are known to degrade when confronted with noisy data, especially when the system is trained only on clean data. In this paper, we show that augmenting training data with sentences containing artificially-introduced grammatical errors can make the system more robust to such errors. In combination with an automatic grammar error correction system, we can recover 1.5 BLEU out of 2.4 BLEU lost due to grammatical errors. We also present a set of Spanish translations of the JFLEG grammar error correction corpus, which allows for testing NMT robustness to real grammatical errors.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. Any repos shown below are weak matches — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Similar Papers

LM-Critic: Language Models for Unsupervised Grammatical Error Correction

86% match1 repo1182021

Evaluation of really good grammatical error correction

83% match1 repo22023

GrammarGPT: Exploring Open-Source LLMs for Native Chinese Grammatical Error Correction with Supervised Fine-Tuning

83% match1 repo1782023

Improving Grammatical Error Correction via Contextual Data Augmentation

83% match1 repo32024

Enhancing Grammatical Error Correction Systems with Explanations

82% match1 repo292023

FlaCGEC: A Chinese Grammatical Error Correction Dataset with Fine-grained Linguistic Annotation

82% match1 repo112023

Cite this paper

@article{anastasopoulos2018neural,
  title  = {Neural Machine Translation of Text from Non-Native Speakers},
  author = {Antonios Anastasopoulos and Alison Lui and Toan Nguyen and David Chiang},
  year   = {2018},
  eprint = {1808.06267},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1808.06267}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1808.06267.svg)](https://codeofpaper.com/paper/1808.06267)

Discussion