MTNT: A Testbed for Machine Translation of Noisy Text

Paul Michel

[0]

Graham Neubig

[0]

EMNLP, pp. 543-553, 2018.

Cited by: 60|Views24

Keywords:

out-of-vocabulary wordsNeural Machine Translationnoisy inputdomain datumsocial mediumMore(13+)

Weibo:

We proposed a new dataset to test Machine Translation models for robustness to the types of noise encountered in natural language on the Internet

Abstract:

Noisy or non-standard input text can cause disastrous mistranslations in most modern Machine Translation (MT) systems, and there has been growing research interest in creating noise-robust MT systems. However, as of yet there are no publicly available parallel corpora of with naturally occurring noisy inputs and translations, and thus pre...More

Code:

Data:

Full Text

View via Publisher