Czech Grammar Agreement Dataset for Evaluation of Language Models

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216224%3A14330%2F16%3A00091975" target="_blank" >RIV/00216224:14330/16:00091975 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Czech Grammar Agreement Dataset for Evaluation of Language Models
Original language description
AGREE is a dataset and task for evaluation of language models based on grammar agreement in Czech. The dataset consists of sentences with marked suffixes of past tense verbs. The task is to choose the right verb suffix which depends on gender, number and animacy of subject. It is challenging for language models because 1) Czech is morphologically rich, 2) it has relatively free word order, 3) high out-of-vocabulary (OOV) ratio, 4) predicate and subject can be far from each other, 5) subjects can be unexpressed and 6) various semantic rules may apply. The task provides a straightforward and easily reproducible way of evaluating language models on a morphologically rich language.
Czech name
—
Czech description
—

Project
<a href="/en/project/7F14047" target="_blank" >7F14047: Harvesting big text data for under-resourced languages</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Publication year
2016
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
RASLAN 2016 Recent Advances in Slavonic Natural Language Processing
ISBN
9788026310952
ISSN
2336-4289
e-ISSN
—
Number of pages
5
Pages from-to
63-67
Publisher name
Tribun EU
Place of publication
Brno
Event location
Karlova studánka
Event date
Jan 1, 2016
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—

Similar results(10)