@inproceedings{wicks-etal-2024-recovering,
title = "Recovering document annotations for sentence-level bitext",
author = "Wicks, Rachel and
Post, Matt and
Koehn, Philipp",
editor = "Ku, Lun-Wei and
Martins, Andre and
Srikumar, Vivek",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2024",
month = aug,
year = "2024",
address = "Bangkok, Thailand",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.findings-acl.589",
doi = "10.18653/v1/2024.findings-acl.589",
pages = "9876--9890",
abstract = "In machine translation, historical models were incapable of handling longer contexts, so the lack of document-level datasets was less noticeable. Now, despite the emergence of long-sequence methods, we remain within a sentence-level paradigm and without data to adequately approach context-aware machine translation. Most large-scale datasets have been processed through a pipeline that discards document-level metadata. In this work, we reconstruct document-level information for three (ParaCrawl, News Commentary, and Europarl) large datasets in German, French, Spanish, Italian, Polish, and Portuguese (paired with English). We then introduce a document-level filtering technique as an alternative to traditional bitext filtering. We present this filtering with analysis to show that this method prefers context-consistent translations rather than those that may have been sentence-level machine translated. Last we train models on these longer contexts and demonstrate improvement in document-level translation without degradation of sentence-level translation. We release our dataset, ParaDocs, and resulting models as a resource to the community.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wicks-etal-2024-recovering">
<titleInfo>
<title>Recovering document annotations for sentence-level bitext</title>
</titleInfo>
<name type="personal">
<namePart type="given">Rachel</namePart>
<namePart type="family">Wicks</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Matt</namePart>
<namePart type="family">Post</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Philipp</namePart>
<namePart type="family">Koehn</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2024</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lun-Wei</namePart>
<namePart type="family">Ku</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andre</namePart>
<namePart type="family">Martins</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vivek</namePart>
<namePart type="family">Srikumar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Bangkok, Thailand</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>In machine translation, historical models were incapable of handling longer contexts, so the lack of document-level datasets was less noticeable. Now, despite the emergence of long-sequence methods, we remain within a sentence-level paradigm and without data to adequately approach context-aware machine translation. Most large-scale datasets have been processed through a pipeline that discards document-level metadata. In this work, we reconstruct document-level information for three (ParaCrawl, News Commentary, and Europarl) large datasets in German, French, Spanish, Italian, Polish, and Portuguese (paired with English). We then introduce a document-level filtering technique as an alternative to traditional bitext filtering. We present this filtering with analysis to show that this method prefers context-consistent translations rather than those that may have been sentence-level machine translated. Last we train models on these longer contexts and demonstrate improvement in document-level translation without degradation of sentence-level translation. We release our dataset, ParaDocs, and resulting models as a resource to the community.</abstract>
<identifier type="citekey">wicks-etal-2024-recovering</identifier>
<identifier type="doi">10.18653/v1/2024.findings-acl.589</identifier>
<location>
<url>https://aclanthology.org/2024.findings-acl.589</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>9876</start>
<end>9890</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Recovering document annotations for sentence-level bitext
%A Wicks, Rachel
%A Post, Matt
%A Koehn, Philipp
%Y Ku, Lun-Wei
%Y Martins, Andre
%Y Srikumar, Vivek
%S Findings of the Association for Computational Linguistics: ACL 2024
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand
%F wicks-etal-2024-recovering
%X In machine translation, historical models were incapable of handling longer contexts, so the lack of document-level datasets was less noticeable. Now, despite the emergence of long-sequence methods, we remain within a sentence-level paradigm and without data to adequately approach context-aware machine translation. Most large-scale datasets have been processed through a pipeline that discards document-level metadata. In this work, we reconstruct document-level information for three (ParaCrawl, News Commentary, and Europarl) large datasets in German, French, Spanish, Italian, Polish, and Portuguese (paired with English). We then introduce a document-level filtering technique as an alternative to traditional bitext filtering. We present this filtering with analysis to show that this method prefers context-consistent translations rather than those that may have been sentence-level machine translated. Last we train models on these longer contexts and demonstrate improvement in document-level translation without degradation of sentence-level translation. We release our dataset, ParaDocs, and resulting models as a resource to the community.
%R 10.18653/v1/2024.findings-acl.589
%U https://aclanthology.org/2024.findings-acl.589
%U https://doi.org/10.18653/v1/2024.findings-acl.589
%P 9876-9890
Markdown (Informal)
[Recovering document annotations for sentence-level bitext](https://aclanthology.org/2024.findings-acl.589) (Wicks et al., Findings 2024)
ACL