English-Czech Systems in WMT19: Document-Level Transformer

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F19%3A10405597" target="_blank" >RIV/00216208:11320/19:10405597 - isvavai.cz</a>

Result on the web

<a href="http://www.statmt.org/wmt19/pdf/53/WMT37.pdf" target="_blank" >http://www.statmt.org/wmt19/pdf/53/WMT37.pdf</a>

DOI - Digital Object Identifier

—

Result language

angličtina

Original language name

English-Czech Systems in WMT19: Document-Level Transformer

Original language description

We describe our NMT systems submitted to the WMT19 shared task in EnglishRIGHTWARDS ARROWCzech news translation. Our systems are based on the Transformer model implemented in either Tensor2Tensor (T2T) or Marian framework. We aimed at improving the adequacy and coherence of translated documents by enlarging the context of the source and target. Instead of translating each sentence independently, we split the document into possibly overlapping multi-sentence segments. In case of the T2T implementation, this &quot;document-level&quot;-trained system achieves a +0.6 BLEU improvement (p &lt; 0.05) relative to the same system applied on isolated sentences. To assess the potential effect document-level models might have on lexical coherence, we performed a semi-automatic analysis, which revealed only a few sentences improved in this aspect. Thus, we cannot draw any conclusions from this week evidence.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

Result was created during the realization of more than one project. More information in the Projects tab.

Continuities

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year

2019

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

Fourth Conference on Machine Translation - Proceedings of the Conference

ISBN

978-1-950737-27-7

ISSN

—

e-ISSN

—

Number of pages

7

Pages from-to

342-348

Publisher name

Association for Computational Linguistics

Place of publication

Stroudsburg, PA, USA

Event location

Firenze, Italy

Event date

Aug 1, 2019

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

—