LIG-CRIStAL System for the WMT17 Automatic Post-Editing Task

Abstract : This paper presents the LIG-CRIStAL submission to the shared Automatic Post-Editing task of WMT 2017. We propose two neural post-editing models: a mono-source model with a task-specific attention mechanism, which performs particularly well in a low-resource scenario; and a chained architecture which makes use of the source sentence to provide extra context. This latter architecture manages to slightly improve our results when more training data is available. We present and discuss our results on two datasets (en-de and de-en) that are made available for the task.
Document type :
Conference papers
Liste complète des métadonnées

Cited literature [18 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01580881
Contributor : Laurent Besacier <>
Submitted on : Sunday, September 3, 2017 - 12:07:53 PM
Last modification on : Thursday, April 4, 2019 - 10:18:05 AM
Document(s) archivé(s) le : Monday, December 11, 2017 - 6:00:24 PM

File

main.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01580881, version 1

Citation

Alexandre Bérard, Olivier Pietquin, Laurent Besacier. LIG-CRIStAL System for the WMT17 Automatic Post-Editing Task. Second conference on machine translation (WMT17) during EMNLP 2017, Sep 2017, Copenhague, Denmark. ⟨hal-01580881⟩

Share

Metrics

Record views

397

Files downloads

179