Training Dataset and Dictionary Sizes Matter in BERT Models: The Case of Baltic Languages

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F22%3A4SGWFY9I" target="_blank" >RIV/00216208:11320/22:4SGWFY9I - isvavai.cz</a>

Result on the web

<a href="https://www.researchgate.net/publication/357201955_Training_dataset_and_dictionary_sizes_matter_in_BERT_models_the_case_of_Baltic_languages" target="_blank" >https://www.researchgate.net/publication/357201955_Training_dataset_and_dictionary_sizes_matter_in_BERT_models_the_case_of_Baltic_languages</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-031-16500-9_14" target="_blank" >10.1007/978-3-031-16500-9_14</a>

Result language

angličtina

Original language name

Training Dataset and Dictionary Sizes Matter in BERT Models: The Case of Baltic Languages

Original language description

Large pretrained masked language models have become state-of-the-art solutions for many NLP problems. While studies have shown that monolingual models produce better results than multilingual models, the training datasets must be sufficiently large. We trained a trilingual LitLat BERT-like model for Lithuanian, Latvian, and English, and a monolingual Est-RoBERTa model for Estonian. We evaluate their performance on four downstream tasks: named entity recognition, dependency parsing, part-of-speech tagging, and word analogy. To analyze the importance of focusing on a single language and the importance of a large training set, we compare created models with existing monolingual and multilingual BERT models for Estonian, Latvian, and Lithuanian. The results show that the newly created LitLat BERT and Est-RoBERTa models improve the results of existing models on all tested tasks in most situations.

Czech name

—

Czech description

—

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)

Project

—

Continuities

—

Publication year

2022

Confidentiality

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

Article name in the collection

Analysis of Images, Social Networks and Texts

ISBN

978-3-031-16500-9

ISSN

—

e-ISSN

—

Number of pages

11

Pages from-to

162-172

Publisher name

Springer International Publishing

Place of publication

—

Event location

Cham

Event date

Jan 1, 2022

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

—