Další formáty:
BibTeX
LaTeX
RIS
@inproceedings{61530,
author = {Signoroni, Edoardo},
address = {Brno},
booktitle = {RASLAN 2023 Recent Advances in Slavonic Natural Language Processing},
keywords = {machine translation;large language models;English;Silesian;evaluation;zero-shot},
howpublished = {tištěná verze "print"},
language = {eng},
location = {Brno},
isbn = {978-80-263-1793-7},
pages = {145-156},
publisher = {Tribun EU},
title = {Fine-Grained Language Relatedness for Zero-Shot Silesian-English Translation},
url = {https://nlp.fi.muni.cz/raslan/raslan23.pdf#page=153},
year = {2023}
}
TY - JOUR
ID - 61530
AU - Signoroni, Edoardo
PY - 2023
TI - Fine-Grained Language Relatedness for Zero-Shot Silesian-English Translation
PB - Tribun EU
CY - Brno
SN - 9788026317937
KW - machine translation;large language models;English;Silesian;evaluation;zero-shot
UR - https://nlp.fi.muni.cz/raslan/raslan23.pdf#page=153
N2 - When parallel corpora are not available to train or fine-tune Machine Translation (MT) systems, one solution is to use data from a related language, and operate in a zero-shot setting. We explore the behaviour and performance of two pre-trained Large Language Models (LLMs) for zero-shot Silesian-English translation, by fine-tuning them on increasingly related languages. Our experiment shows that using data from related languages generally improves the zero-shot translation performance for our language pair, but the optimal fine-grained choice inside the Slavic language family is non-trivial and depends on the model characteristics.
ER -
SIGNORONI, Edoardo. Fine-Grained Language Relatedness for Zero-Shot Silesian-English Translation. In \textit{RASLAN 2023 Recent Advances in Slavonic Natural Language Processing}. Brno: Tribun EU, 2023, s.~145-156. ISBN~978-80-263-1793-7.
|
