Context-Aware Monolingual Repair for Neural Machine Translation
| Authors | |
|---|---|
| Publication date | 2019 |
| Host editors |
|
| Book title | 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint Conference on Natural Language Processing |
| Book subtitle | EMNLP-IJCNLP 2019 : proceedings of the conference : November 3-7, 2019, Hong Kong, China |
| ISBN (electronic) |
|
| Event | 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing |
| Pages (from-to) | 877-886 |
| Publisher | Stroudsburg, PA: The Association for Computational Linguistics |
| Organisations |
|
| Abstract |
Modern sentence-level NMT systems often produce plausible translations of isolated sentences. However, when put in context, these translations may end up being inconsistent with each other. We propose a monolingual DocRepair model to correct inconsistencies between sentence-level translations. DocRepair performs automatic post-editing on a sequence of sentence-level translations, refining translations of sentences in context of each other. For training, the DocRepair model requires only monolingual document-level data in the target language. It is trained as a monolingual sequence-to-sequence model that maps inconsistent groups of sentences into consistent ones. The consistent groups come from the original training data; the inconsistent groups are obtained by sampling round-trip translations for each isolated sentence. We show that this approach successfully imitates inconsistencies we aim to fix: using contrastive evaluation, we show large improvements in the translation of several contextual phenomena in an English-Russian translation task, as well as improvements in the BLEU score. We also conduct a human evaluation and show a strong preference of the annotators to corrected translations over the baseline ones. Moreover, we analyze which discourse phenomena are hard to capture using monolingual data only.
|
| Document type | Conference contribution |
| Language | English |
| Published at | https://doi.org/10.18653/v1/D19-1081 |
| Other links | https://github.com/lena-voita/good-translation-wrong-in-context |
| Downloads |
D19-1081
(Final published version)
|
| Permalink to this page | |