BUT-FIT at SemEval-2020 Task 4: Multilingual commonsense

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F20%3APU138894" target="_blank" >RIV/00216305:26230/20:PU138894 - isvavai.cz</a>

Výsledek na webu

<a href="https://www.aclweb.org/anthology/2020.semeval-1.46/" target="_blank" >https://www.aclweb.org/anthology/2020.semeval-1.46/</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

BUT-FIT at SemEval-2020 Task 4: Multilingual commonsense

Popis výsledku v původním jazyce

We participated in all three subtasks. In subtasks A and B, our submissions are based on pretrained language representation models (namely ALBERT) and data augmentation. We experimented with solving the task for another language, Czech, by means of multilingual models and machine translated dataset, or translated model inputs. We show that with a strong machine translation system, our system can be used in another language with a small accuracy loss. In subtask C, our submission, which is based on pretrained sequence-to-sequence model (BART), ranked 1st in BLEU score ranking, however, we show that the correlation between BLEU and human evaluation, in which our submission ended up 4th, is low. We analyse the metrics used in the evaluation and we propose an additional score based on model from subtask B, which correlates well with our manual ranking, as well as reranking method based on the same principle. We performed an error and dataset analysis for all subtasks and we present our findings.

Název v anglickém jazyce

BUT-FIT at SemEval-2020 Task 4: Multilingual commonsense

Popis výsledku anglicky

We participated in all three subtasks. In subtasks A and B, our submissions are based on pretrained language representation models (namely ALBERT) and data augmentation. We experimented with solving the task for another language, Czech, by means of multilingual models and machine translated dataset, or translated model inputs. We show that with a strong machine translation system, our system can be used in another language with a small accuracy loss. In subtask C, our submission, which is based on pretrained sequence-to-sequence model (BART), ranked 1st in BLEU score ranking, however, we show that the correlation between BLEU and human evaluation, in which our submission ended up 4th, is low. We analyse the metrics used in the evaluation and we propose an additional score based on model from subtask B, which correlates well with our manual ranking, as well as reranking method based on the same principle. We performed an error and dataset analysis for all subtasks and we present our findings.

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/LTC18054" target="_blank" >LTC18054: Pokročilé sémantické obohacování vícejazyčných kolekcí literárních textů</a><br>

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

Proceedings of the Fourteenth Workshop on Semantic Evaluation

ISBN

978-1-952148-31-6

ISSN

—

e-ISSN

—

Počet stran výsledku

16

Strana od-do

374-390

Název nakladatele

Association for Computational Linguistics

Místo vydání

Barcelona

Místo konání akce

Barcelona (online)

Datum konání akce

8. 12. 2020

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—