Leveraging Auxiliary Domain Parallel Data in Intermediate Task Fine-tuning for Low-resource Translation

NMT systems trained on Pre-trained Multilingual Sequence-Sequence (PMSS) models flounder when sufficient amounts of parallel data is not available for fine-tuning. This specifically holds for languages missing/under-represented in these models. The problem gets aggravated when the data comes from di...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2023-06
Main Authors: Nayak, Shravan, Ranathunga, Surangika, Thillainathan, Sarubi, Hung, Rikki, Rinaldi, Anthony, Wang, Yining, Mackey, Jonah, Ho, Andrew, En-Shiun Annie Lee
Format: Article
Language:English
Subjects:
Languages
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Nayak, Shravan
Ranathunga, Surangika
Thillainathan, Sarubi
Hung, Rikki
Rinaldi, Anthony
Wang, Yining
Mackey, Jonah
Ho, Andrew
En-Shiun Annie Lee
description NMT systems trained on Pre-trained Multilingual Sequence-Sequence (PMSS) models flounder when sufficient amounts of parallel data is not available for fine-tuning. This specifically holds for languages missing/under-represented in these models. The problem gets aggravated when the data comes from different domains. In this paper, we show that intermediate-task fine-tuning (ITFT) of PMSS models is extremely beneficial for domain-specific NMT, especially when target domain data is limited/unavailable and the considered languages are missing or under-represented in the PMSS model. We quantify the domain-specific results variations using a domain-divergence test, and show that ITFT can mitigate the impact of domain divergence to some extent.
format article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2822566501</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2822566501</sourcerecordid><originalsourceid>FETCH-proquest_journals_28225665013</originalsourceid><addsrcrecordid>eNqNi80KgkAURocgSMp3GGgt6NiY28ikwEUL93Gpq4yNM3Vn7OftM-gBWn1wvnMmLBBpmkT5SogZC53r4jgW2VpImQasrfCBBK0yLd8ML6UV0JsXtgdl-BEItEbNC_DAR3AwHqnHiwKPvAZ35aUyGPnBfPvGEq_sMyJ0dqDzaBAYp8EraxZs2oB2GP52zpblrt7uoxvZ-4DOn7qxMeN1ErkQMstknKT_WR_9R0d3</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2822566501</pqid></control><display><type>article</type><title>Leveraging Auxiliary Domain Parallel Data in Intermediate Task Fine-tuning for Low-resource Translation</title><source>Publicly Available Content Database</source><creator>Nayak, Shravan ; Ranathunga, Surangika ; Thillainathan, Sarubi ; Hung, Rikki ; Rinaldi, Anthony ; Wang, Yining ; Mackey, Jonah ; Ho, Andrew ; En-Shiun Annie Lee</creator><creatorcontrib>Nayak, Shravan ; Ranathunga, Surangika ; Thillainathan, Sarubi ; Hung, Rikki ; Rinaldi, Anthony ; Wang, Yining ; Mackey, Jonah ; Ho, Andrew ; En-Shiun Annie Lee</creatorcontrib><description>NMT systems trained on Pre-trained Multilingual Sequence-Sequence (PMSS) models flounder when sufficient amounts of parallel data is not available for fine-tuning. This specifically holds for languages missing/under-represented in these models. The problem gets aggravated when the data comes from different domains. In this paper, we show that intermediate-task fine-tuning (ITFT) of PMSS models is extremely beneficial for domain-specific NMT, especially when target domain data is limited/unavailable and the considered languages are missing or under-represented in the PMSS model. We quantify the domain-specific results variations using a domain-divergence test, and show that ITFT can mitigate the impact of domain divergence to some extent.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Languages</subject><ispartof>arXiv.org, 2023-06</ispartof><rights>2023. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://www.proquest.com/docview/2822566501?pq-origsite=primo$$EHTML$$P50$$Gproquest$$Hfree_for_read</linktohtml><link.rule.ids>780,784,25753,37012,44590</link.rule.ids></links><search><creatorcontrib>Nayak, Shravan</creatorcontrib><creatorcontrib>Ranathunga, Surangika</creatorcontrib><creatorcontrib>Thillainathan, Sarubi</creatorcontrib><creatorcontrib>Hung, Rikki</creatorcontrib><creatorcontrib>Rinaldi, Anthony</creatorcontrib><creatorcontrib>Wang, Yining</creatorcontrib><creatorcontrib>Mackey, Jonah</creatorcontrib><creatorcontrib>Ho, Andrew</creatorcontrib><creatorcontrib>En-Shiun Annie Lee</creatorcontrib><title>Leveraging Auxiliary Domain Parallel Data in Intermediate Task Fine-tuning for Low-resource Translation</title><title>arXiv.org</title><description>NMT systems trained on Pre-trained Multilingual Sequence-Sequence (PMSS) models flounder when sufficient amounts of parallel data is not available for fine-tuning. This specifically holds for languages missing/under-represented in these models. The problem gets aggravated when the data comes from different domains. In this paper, we show that intermediate-task fine-tuning (ITFT) of PMSS models is extremely beneficial for domain-specific NMT, especially when target domain data is limited/unavailable and the considered languages are missing or under-represented in the PMSS model. We quantify the domain-specific results variations using a domain-divergence test, and show that ITFT can mitigate the impact of domain divergence to some extent.</description><subject>Languages</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>PIMPY</sourceid><recordid>eNqNi80KgkAURocgSMp3GGgt6NiY28ikwEUL93Gpq4yNM3Vn7OftM-gBWn1wvnMmLBBpmkT5SogZC53r4jgW2VpImQasrfCBBK0yLd8ML6UV0JsXtgdl-BEItEbNC_DAR3AwHqnHiwKPvAZ35aUyGPnBfPvGEq_sMyJ0dqDzaBAYp8EraxZs2oB2GP52zpblrt7uoxvZ-4DOn7qxMeN1ErkQMstknKT_WR_9R0d3</recordid><startdate>20230602</startdate><enddate>20230602</enddate><creator>Nayak, Shravan</creator><creator>Ranathunga, Surangika</creator><creator>Thillainathan, Sarubi</creator><creator>Hung, Rikki</creator><creator>Rinaldi, Anthony</creator><creator>Wang, Yining</creator><creator>Mackey, Jonah</creator><creator>Ho, Andrew</creator><creator>En-Shiun Annie Lee</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20230602</creationdate><title>Leveraging Auxiliary Domain Parallel Data in Intermediate Task Fine-tuning for Low-resource Translation</title><author>Nayak, Shravan ; Ranathunga, Surangika ; Thillainathan, Sarubi ; Hung, Rikki ; Rinaldi, Anthony ; Wang, Yining ; Mackey, Jonah ; Ho, Andrew ; En-Shiun Annie Lee</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_28225665013</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Languages</topic><toplevel>online_resources</toplevel><creatorcontrib>Nayak, Shravan</creatorcontrib><creatorcontrib>Ranathunga, Surangika</creatorcontrib><creatorcontrib>Thillainathan, Sarubi</creatorcontrib><creatorcontrib>Hung, Rikki</creatorcontrib><creatorcontrib>Rinaldi, Anthony</creatorcontrib><creatorcontrib>Wang, Yining</creatorcontrib><creatorcontrib>Mackey, Jonah</creatorcontrib><creatorcontrib>Ho, Andrew</creatorcontrib><creatorcontrib>En-Shiun Annie Lee</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni)</collection><collection>ProQuest Central</collection><collection>ProQuest Central Essentials</collection><collection>AUTh Library subscriptions: ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Nayak, Shravan</au><au>Ranathunga, Surangika</au><au>Thillainathan, Sarubi</au><au>Hung, Rikki</au><au>Rinaldi, Anthony</au><au>Wang, Yining</au><au>Mackey, Jonah</au><au>Ho, Andrew</au><au>En-Shiun Annie Lee</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Leveraging Auxiliary Domain Parallel Data in Intermediate Task Fine-tuning for Low-resource Translation</atitle><jtitle>arXiv.org</jtitle><date>2023-06-02</date><risdate>2023</risdate><eissn>2331-8422</eissn><abstract>NMT systems trained on Pre-trained Multilingual Sequence-Sequence (PMSS) models flounder when sufficient amounts of parallel data is not available for fine-tuning. This specifically holds for languages missing/under-represented in these models. The problem gets aggravated when the data comes from different domains. In this paper, we show that intermediate-task fine-tuning (ITFT) of PMSS models is extremely beneficial for domain-specific NMT, especially when target domain data is limited/unavailable and the considered languages are missing or under-represented in the PMSS model. We quantify the domain-specific results variations using a domain-divergence test, and show that ITFT can mitigate the impact of domain divergence to some extent.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2023-06
issn 2331-8422
language eng
recordid cdi_proquest_journals_2822566501
source Publicly Available Content Database
subjects Languages
title Leveraging Auxiliary Domain Parallel Data in Intermediate Task Fine-tuning for Low-resource Translation
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-01T23%3A24%3A15IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Leveraging%20Auxiliary%20Domain%20Parallel%20Data%20in%20Intermediate%20Task%20Fine-tuning%20for%20Low-resource%20Translation&rft.jtitle=arXiv.org&rft.au=Nayak,%20Shravan&rft.date=2023-06-02&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2822566501%3C/proquest%3E%3Cgrp_id%3Ecdi_FETCH-proquest_journals_28225665013%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2822566501&rft_id=info:pmid/&rfr_iscdi=true