Subword-based Cross-lingual Transfer of Embeddings from Hindi to Marathi and Nepali

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F22%3A10456922" target="_blank" >RIV/00216208:11320/22:10456922 - isvavai.cz</a>

Výsledek na webu

<a href="https://aclanthology.org/2022.sigmorphon-1.pdf" target="_blank" >https://aclanthology.org/2022.sigmorphon-1.pdf</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Subword-based Cross-lingual Transfer of Embeddings from Hindi to Marathi and Nepali

Popis výsledku v původním jazyce

Word embeddings are growing to be a crucial resource in the field of NLP for any language. This work introduces a novel technique for static subword embeddings transfer for Indic languages from a relatively higher resource language to a genealogically related low resource language. We primarily work with Hindi-Marathi, simulating a low-resource scenario for Marathi, and confirm observed trends on Nepali. We demonstrate the consistent benefits of unsupervised morphemic segmentation on both source and target sides over the treatment performed by fastText. Our best-performing approach uses an EM-style approach to learning bilingual subword embeddings; we also show, for the first time, that a trivial &quot;copy- and-paste&quot; embeddings transfer based on even perfect bilingual lexicons is inadequate in capturing language-specific relationships. We find that our approach substantially outperforms the fastText baselines for both Marathi and Nepali on the Word Similarity task as well as WordNet-Based Synonymy Tests;

Název v anglickém jazyce

Subword-based Cross-lingual Transfer of Embeddings from Hindi to Marathi and Nepali

Popis výsledku anglicky

Word embeddings are growing to be a crucial resource in the field of NLP for any language. This work introduces a novel technique for static subword embeddings transfer for Indic languages from a relatively higher resource language to a genealogically related low resource language. We primarily work with Hindi-Marathi, simulating a low-resource scenario for Marathi, and confirm observed trends on Nepali. We demonstrate the consistent benefits of unsupervised morphemic segmentation on both source and target sides over the treatment performed by fastText. Our best-performing approach uses an EM-style approach to learning bilingual subword embeddings; we also show, for the first time, that a trivial &quot;copy- and-paste&quot; embeddings transfer based on even perfect bilingual lexicons is inadequate in capturing language-specific relationships. We find that our approach substantially outperforms the fastText baselines for both Marathi and Nepali on the Word Similarity task as well as WordNet-Based Synonymy Tests;

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2022

Kód důvěrnosti údajů

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

Název statě ve sborníku

19th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology

ISBN

978-1-955917-82-7

ISSN

—

e-ISSN

—

Počet stran výsledku

11

Strana od-do

61-71

Název nakladatele

Association for Computational Linguistics

Místo vydání

Stroudsburg, PA, USA

Místo konání akce

Seattle, WA, USA

Datum konání akce

14. 7. 2022

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—