HFT: High Frequency Tokens for Low-Resource NMT

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216224%3A14330%2F22%3A00127008" target="_blank" >RIV/00216224:14330/22:00127008 - isvavai.cz</a>
Result on the web
<a href="https://aclanthology.org/2022.loresmt-1.8" target="_blank" >https://aclanthology.org/2022.loresmt-1.8</a>
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
HFT: High Frequency Tokens for Low-Resource NMT
Original language description
Tokenization has been shown to impact the quality of downstream tasks, such as Neural Machine Translation (NMT), which is susceptible to out-of-vocabulary words and low frequency training data. Current state-of-the-art algorithms have been helpful in addressing the issues of out-of-vocabulary words, bigger vocabulary sizes and token frequency by implementing subword segmentation. We argue, however, that there is still room for improvement, in particular regarding low-frequency tokens in the training data. In this paper, we present “High Frequency Tokenizer”, or HFT, a new language-independent subword segmentation algorithm that addresses this issue. We also propose a new metric to measure the frequency coverage of a tokenizer’s vocabulary, based on a frequency rank weighted average of the frequency values of its items. We experiment with a diverse set of language corpora, vocabulary sizes, and writing systems and report improvements on both frequency statistics and on the average length of the output. We also observe a positive impact on downstream NMT.
Czech name
—
Czech description
—

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)

Publication year
2022
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
Proceedings of the Fifth Workshop on Technologies for Machine Translation of Low-Resource Languages (LoResMT 2022)
ISBN
—
ISSN
2951-2093
e-ISSN
—
Number of pages
8
Pages from-to
56-63
Publisher name
Association for Computational Linguistics
Place of publication
Gyeongju, Republic of Korea
Event location
Gyeongju, Republic of Korea
Event date
Oct 16, 2022
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—

Similar results(10)