Online Versus Offline NMT Quality: An In-depth Analysis on English–German and German–English - Archive ouverte HAL Access content directly
Conference Papers Year : 2020

Online Versus Offline NMT Quality: An In-depth Analysis on English–German and German–English

Abstract

We conduct in this work an evaluation study comparing offline and online neural machine translation architectures. Two sequence-to-sequence models: convolutional Pervasive Attention (Elbayad et al., 2018) and attention-based Transformer (Vaswani et al., 2017) are considered. We investigate, for both architectures, the impact of online decoding constraints on the translation quality through a carefully designed human evaluation on English-German and German-English language pairs, the latter being particularly sensitive to latency constraints. The evaluation results allow us to identify the strengths and shortcomings of each model when we shift to the online setup.
Fichier principal
Vignette du fichier
Error_Analysis__COLING2020_-3.pdf (438.47 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-02991539 , version 1 (06-11-2020)

Identifiers

Cite

Maha Elbayad, Michael Ustaszewski, Emmanuelle Esperança-Rodier, Francis Brunet-Manquat, Jakob Verbeek, et al.. Online Versus Offline NMT Quality: An In-depth Analysis on English–German and German–English. COLING 2020 - 28th International Conference on Computational Linguistics, Dec 2020, Virtual, Spain. pp.5047-5058, ⟨10.18653/v1/2020.coling-main.443⟩. ⟨hal-02991539⟩
134 View
177 Download

Altmetric

Share

Gmail Facebook X LinkedIn More