@inproceedings{okabe-fraser-2025-bilingual,
title = "Bilingual Sentence Mining for Low-Resource Languages: a Case Study on Upper and {L}ower {S}orbian",
author = "Okabe, Shu and
Fraser, Alexander",
editor = "Lachler, Jordan and
Agyapong, Godfred and
Arppe, Antti and
Moeller, Sarah and
Chaudhary, Aditi and
Rijhwani, Shruti and
Rosenblum, Daisy",
booktitle = "Proceedings of the Eight Workshop on the Use of Computational Methods in the Study of Endangered Languages",
month = mar,
year = "2025",
address = "Honolulu, Hawaii, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.computel-main.2/",
pages = "11--19",
abstract = "Parallel sentence mining is crucial for down- stream tasks such as Machine Translation, especially for low-resource languages, where such resources are scarce. In this context, we apply a pipeline approach with contextual embeddings on two endangered Slavic languages spoken in Germany, Upper and Lower Sorbian, to evaluate mining quality. To this end, we compare off-the-shelf multilingual language models and word encoders pre-trained on Upper Sorbian to understand their impact on sentence mining. Moreover, to filter out irrelevant pairs, we experiment with a post-processing of mined sentences through an unsupervised word aligner based on word embeddings. We observe the usefulness of additional pre-training in Upper Sorbian, which leads to direct improvements when mining the same language but also its related language, Lower Sorbian."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="okabe-fraser-2025-bilingual">
<titleInfo>
<title>Bilingual Sentence Mining for Low-Resource Languages: a Case Study on Upper and Lower Sorbian</title>
</titleInfo>
<name type="personal">
<namePart type="given">Shu</namePart>
<namePart type="family">Okabe</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alexander</namePart>
<namePart type="family">Fraser</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-03</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Eight Workshop on the Use of Computational Methods in the Study of Endangered Languages</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jordan</namePart>
<namePart type="family">Lachler</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Godfred</namePart>
<namePart type="family">Agyapong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Antti</namePart>
<namePart type="family">Arppe</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sarah</namePart>
<namePart type="family">Moeller</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aditi</namePart>
<namePart type="family">Chaudhary</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shruti</namePart>
<namePart type="family">Rijhwani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daisy</namePart>
<namePart type="family">Rosenblum</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Honolulu, Hawaii, USA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Parallel sentence mining is crucial for down- stream tasks such as Machine Translation, especially for low-resource languages, where such resources are scarce. In this context, we apply a pipeline approach with contextual embeddings on two endangered Slavic languages spoken in Germany, Upper and Lower Sorbian, to evaluate mining quality. To this end, we compare off-the-shelf multilingual language models and word encoders pre-trained on Upper Sorbian to understand their impact on sentence mining. Moreover, to filter out irrelevant pairs, we experiment with a post-processing of mined sentences through an unsupervised word aligner based on word embeddings. We observe the usefulness of additional pre-training in Upper Sorbian, which leads to direct improvements when mining the same language but also its related language, Lower Sorbian.</abstract>
<identifier type="citekey">okabe-fraser-2025-bilingual</identifier>
<location>
<url>https://aclanthology.org/2025.computel-main.2/</url>
</location>
<part>
<date>2025-03</date>
<extent unit="page">
<start>11</start>
<end>19</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Bilingual Sentence Mining for Low-Resource Languages: a Case Study on Upper and Lower Sorbian
%A Okabe, Shu
%A Fraser, Alexander
%Y Lachler, Jordan
%Y Agyapong, Godfred
%Y Arppe, Antti
%Y Moeller, Sarah
%Y Chaudhary, Aditi
%Y Rijhwani, Shruti
%Y Rosenblum, Daisy
%S Proceedings of the Eight Workshop on the Use of Computational Methods in the Study of Endangered Languages
%D 2025
%8 March
%I Association for Computational Linguistics
%C Honolulu, Hawaii, USA
%F okabe-fraser-2025-bilingual
%X Parallel sentence mining is crucial for down- stream tasks such as Machine Translation, especially for low-resource languages, where such resources are scarce. In this context, we apply a pipeline approach with contextual embeddings on two endangered Slavic languages spoken in Germany, Upper and Lower Sorbian, to evaluate mining quality. To this end, we compare off-the-shelf multilingual language models and word encoders pre-trained on Upper Sorbian to understand their impact on sentence mining. Moreover, to filter out irrelevant pairs, we experiment with a post-processing of mined sentences through an unsupervised word aligner based on word embeddings. We observe the usefulness of additional pre-training in Upper Sorbian, which leads to direct improvements when mining the same language but also its related language, Lower Sorbian.
%U https://aclanthology.org/2025.computel-main.2/
%P 11-19
Markdown (Informal)
[Bilingual Sentence Mining for Low-Resource Languages: a Case Study on Upper and Lower Sorbian](https://aclanthology.org/2025.computel-main.2/) (Okabe & Fraser, ComputEL 2025)
ACL