Reading Comprehension in Czech via Machine Translation and Cross-lingual Transfer

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F20%3A10415122" target="_blank" >RIV/00216208:11320/20:10415122 - isvavai.cz</a>

Nalezeny alternativní kódy

RIV/00216208:11320/20:10424477

Výsledek na webu

<a href="https://doi.org/10.1007/978-3-030-58323-1_18" target="_blank" >https://doi.org/10.1007/978-3-030-58323-1_18</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-030-58323-1_18" target="_blank" >10.1007/978-3-030-58323-1_18</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Reading Comprehension in Czech via Machine Translation and Cross-lingual Transfer

Popis výsledku v původním jazyce

Reading comprehension is a well studied task, with huge training datasets in English. This work focuses on building reading comprehension systems for Czech, without requiring any manually annotated Czech training data. First of all, we automatically translated SQuAD 1.1 and SQuAD 2.0 datasets to Czech to create training and development data, which we release at http://hdl.handle.net/11234/1-3249. We then trained and evaluated several BERT and XLM-RoBERTa baseline models. However, our main focus lies in cross-lingual transfer models. We report that a XLM-RoBERTa model trained on English data and evaluated on Czech achieves very competitive performance, only approximately 2% points worse than a model trained on the translated Czech data. This result is extremely good, considering the fact that the model has not seen any Czech data during training. The cross-lingual transfer approach is very flexible and provides a reading comprehension in any language, for which we have enough monolingual raw texts.

Název v anglickém jazyce

Reading Comprehension in Czech via Machine Translation and Cross-lingual Transfer

Popis výsledku anglicky

Reading comprehension is a well studied task, with huge training datasets in English. This work focuses on building reading comprehension systems for Czech, without requiring any manually annotated Czech training data. First of all, we automatically translated SQuAD 1.1 and SQuAD 2.0 datasets to Czech to create training and development data, which we release at http://hdl.handle.net/11234/1-3249. We then trained and evaluated several BERT and XLM-RoBERTa baseline models. However, our main focus lies in cross-lingual transfer models. We report that a XLM-RoBERTa model trained on English data and evaluated on Czech achieves very competitive performance, only approximately 2% points worse than a model trained on the translated Czech data. This result is extremely good, considering the fact that the model has not seen any Czech data during training. The cross-lingual transfer approach is very flexible and provides a reading comprehension in any language, for which we have enough monolingual raw texts.

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

Výsledek vznikl pri realizaci vícero projektů. Více informací v záložce Projekty.

Návaznosti

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

Rok uplatnění

2020

Kód důvěrnosti údajů

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

Název statě ve sborníku

Text, Speech, and Dialogue. TSD 2020

ISBN

978-3-030-58322-4

ISSN

0302-9743

e-ISSN

—

Počet stran výsledku

9

Strana od-do

171-179

Název nakladatele

Springer

Místo vydání

Cham

Místo konání akce

Brno

Datum konání akce

8. 9. 2020

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—