Translating Short Segments with NMT: A Case Study in English-to-Hindi

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F18%3A10390198" target="_blank" >RIV/00216208:11320/18:10390198 - isvavai.cz</a>

Výsledek na webu

<a href="http://rua.ua.es/dspace/handle/10045/76083" target="_blank" >http://rua.ua.es/dspace/handle/10045/76083</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Translating Short Segments with NMT: A Case Study in English-to-Hindi

Popis výsledku v původním jazyce

This paper presents a case study in translating short image captions of the Visual Genome dataset from English into Hindi using out-of-domain data sets of varying size. We experiment with three NMT models: the shallow and deep sequence-to-sequence and the Transformer model as implemented in Marian toolkit. Phrase-based Moses serves as the baseline. The results indicate that the Transformer model outperforms others in the large data setting in a number of automatic metrics and manual evaluation, and it also produces the fewest truncated sentences. Transformer training is however very sensitive to the hyperparameters, so it requires more experimenting. The deep sequence-to-sequence model produced more flawless outputs in the small data setting and it was generally more stable, at the cost of more training iterations.

Název v anglickém jazyce

Translating Short Segments with NMT: A Case Study in English-to-Hindi

Popis výsledku anglicky

This paper presents a case study in translating short image captions of the Visual Genome dataset from English into Hindi using out-of-domain data sets of varying size. We experiment with three NMT models: the shallow and deep sequence-to-sequence and the Transformer model as implemented in Marian toolkit. Phrase-based Moses serves as the baseline. The results indicate that the Transformer model outperforms others in the large data setting in a number of automatic metrics and manual evaluation, and it also produces the fewest truncated sentences. Transformer training is however very sensitive to the hyperparameters, so it requires more experimenting. The deep sequence-to-sequence model produced more flawless outputs in the small data setting and it was generally more stable, at the cost of more training iterations.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

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

Projekt

<a href="/cs/project/LM2015071" target="_blank" >LM2015071: Jazyková výzkumná infrastruktura v České republice</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2018

Kód důvěrnosti údajů

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

Název statě ve sborníku

Proceedings of the 21st Annual Conference of the European Association for Machine Translation (2018)

ISBN

978-84-09-01901-4

ISSN

—

e-ISSN

neuvedeno

Počet stran výsledku

392

Strana od-do

1-392

Název nakladatele

European Association for Machine Translation

Místo vydání

Allschwil, Switzerland

Místo konání akce

Alicante, Spain

Datum konání akce

28. 5. 2018

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—