Results of the WMT20 Metrics Shared Task

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F20%3A10424478" target="_blank" >RIV/00216208:11320/20:10424478 - isvavai.cz</a>

Výsledek na webu

<a href="http://www.statmt.org/wmt20/pdf/2020.wmt-1.77.pdf" target="_blank" >http://www.statmt.org/wmt20/pdf/2020.wmt-1.77.pdf</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Results of the WMT20 Metrics Shared Task

Popis výsledku v původním jazyce

This paper presents the results of the WMT20 Metrics Shared Task. Participants were asked to score the outputs of the translation systems competing in the WMT20 News Translation Task with automatic metrics. Ten research groups submitted 27 metrics, four of which are reference-less &quot;metrics&quot;. In addition, we computed five baseline metrics, including SENT BLEU, BLEU, TER and CHR F using the SacreBLEU scorer. All metrics were evaluated on how well they correlate at the system-, document- and segment-level with the WMT20 official human scores. We present an extensive analysis on influence of reference translations on metric reliability, how well automatic metrics score human trans- lations, and we also flag major discrepancies between metric and human scores when eval- uating MT systems. Finally, we investigate whether we can use automatic metrics to flag incorrect human ratings.

Název v anglickém jazyce

Results of the WMT20 Metrics Shared Task

Popis výsledku anglicky

This paper presents the results of the WMT20 Metrics Shared Task. Participants were asked to score the outputs of the translation systems competing in the WMT20 News Translation Task with automatic metrics. Ten research groups submitted 27 metrics, four of which are reference-less &quot;metrics&quot;. In addition, we computed five baseline metrics, including SENT BLEU, BLEU, TER and CHR F using the SacreBLEU scorer. All metrics were evaluated on how well they correlate at the system-, document- and segment-level with the WMT20 official human scores. We present an extensive analysis on influence of reference translations on metric reliability, how well automatic metrics score human trans- lations, and we also flag major discrepancies between metric and human scores when eval- uating MT systems. Finally, we investigate whether we can use automatic metrics to flag incorrect human ratings.

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/GX19-26934X" target="_blank" >GX19-26934X: Neuronové reprezentace v multimodálním a mnohojazyčném modelování</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

Fifth Conference on Machine Translation - Proceedings of the Conference

ISBN

978-1-948087-81-0

ISSN

—

e-ISSN

—

Počet stran výsledku

38

Strana od-do

688-725

Název nakladatele

Association for Computational Linguistics

Místo vydání

Stroudsburg, PA, USA

Místo konání akce

Online

Datum konání akce

19. 11. 2020

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—