UDPipe at EvaLatin 2020: Contextualized Embeddings and Treebank Embeddings
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F20%3A10424502" target="_blank" >RIV/00216208:11320/20:10424502 - isvavai.cz</a>
Result on the web
<a href="https://www.aclweb.org/anthology/2020.lt4hala-1.20/" target="_blank" >https://www.aclweb.org/anthology/2020.lt4hala-1.20/</a>
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
UDPipe at EvaLatin 2020: Contextualized Embeddings and Treebank Embeddings
Original language description
We present our contribution to the EvaLatin shared task, which is the first evaluation campaign devoted to the evaluation of NLP tools for Latin. We submitted a system based on UDPipe 2.0, one of the winners of the CoNLL 2018 Shared Task, The 2018 Shared Task on Extrinsic Parser Evaluation and SIGMORPHON 2019 Shared Task. Our system places first by a wide margin both in lemmatization and POS tagging in the open modality, where additional supervised data is allowed, in which case we utilize all Universal Dependency Latin treebanks. In the closed modality, where only the EvaLatin training data is allowed, our system achieves the best performance in lemmatization and in classical subtask of POS tagging, while reaching second place in cross-genre and cross-time settings. In the ablation experiments, we also evaluate the influence of BERT and XLM-RoBERTa contextualized embeddings, and the treebank encodings of the different flavors of Latin treebanks.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
—
OECD FORD branch
10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
Result continuities
Project
Result was created during the realization of more than one project. More information in the Projects tab.
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)
Others
Publication year
2020
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
Proceedings of LT4HALA 2020 - 1st Workshop on Language Technologies for Historical and Ancient Languages
ISBN
979-10-95546-53-5
ISSN
—
e-ISSN
—
Number of pages
6
Pages from-to
124-129
Publisher name
European Language Resources Association (ELRA)
Place of publication
Marseille, France
Event location
Marseille, France
Event date
May 12, 2020
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—