@inproceedings{gibert-etal-2025-mind,
title = "Mind the Gap: {Diverse} {NMT} Models for Resource-Constrained Environments",
author = {Gibert, Ona de and
O{'}Brien, Dayy{\'a}n and
Vari{\v{s}}, Du{\v{s}}an and
Tiedemann, J{\"o}rg},
editor = "Johansson, Richard and
Stymne, Sara",
booktitle = "Proceedings of the Joint 25th Nordic Conference on Computational Linguistics and 11th Baltic Conference on Human Language Technologies (NoDaLiDa/Baltic-HLT 2025)",
month = mar,
year = "2025",
address = "Tallinn, Estonia",
publisher = "University of Tartu Library",
url = "https://aclanthology.org/2025.nodalida-1.21/",
pages = "209--216",
ISBN = "978-9908-53-109-0",
abstract = "We present fast Neural Machine Translation models for 17 diverse languages, developed using Sequence-level Knowledge Distillation. Our selected languages span multiple language families and scripts, including low-resource languages. The distilled models achieve comparable performance while being 10x times faster than transformer-base and 35x times faster than transformer-big architectures. Our experiments reveal that teacher model quality and capacity strongly influence the distillation success, as well as the language script. We also explore the effectiveness of multilingual students. We release publicly our code and models in our Github repository: anonymised."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="gibert-etal-2025-mind">
<titleInfo>
<title>Mind the Gap: Diverse NMT Models for Resource-Constrained Environments</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ona</namePart>
<namePart type="given">de</namePart>
<namePart type="family">Gibert</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dayyán</namePart>
<namePart type="family">O’Brien</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dušan</namePart>
<namePart type="family">Variš</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jörg</namePart>
<namePart type="family">Tiedemann</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 Joint 25th Nordic Conference on Computational Linguistics and 11th Baltic Conference on Human Language Technologies (NoDaLiDa/Baltic-HLT 2025)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Richard</namePart>
<namePart type="family">Johansson</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sara</namePart>
<namePart type="family">Stymne</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>University of Tartu Library</publisher>
<place>
<placeTerm type="text">Tallinn, Estonia</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">978-9908-53-109-0</identifier>
</relatedItem>
<abstract>We present fast Neural Machine Translation models for 17 diverse languages, developed using Sequence-level Knowledge Distillation. Our selected languages span multiple language families and scripts, including low-resource languages. The distilled models achieve comparable performance while being 10x times faster than transformer-base and 35x times faster than transformer-big architectures. Our experiments reveal that teacher model quality and capacity strongly influence the distillation success, as well as the language script. We also explore the effectiveness of multilingual students. We release publicly our code and models in our Github repository: anonymised.</abstract>
<identifier type="citekey">gibert-etal-2025-mind</identifier>
<location>
<url>https://aclanthology.org/2025.nodalida-1.21/</url>
</location>
<part>
<date>2025-03</date>
<extent unit="page">
<start>209</start>
<end>216</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Mind the Gap: Diverse NMT Models for Resource-Constrained Environments
%A Gibert, Ona de
%A O’Brien, Dayyán
%A Variš, Dušan
%A Tiedemann, Jörg
%Y Johansson, Richard
%Y Stymne, Sara
%S Proceedings of the Joint 25th Nordic Conference on Computational Linguistics and 11th Baltic Conference on Human Language Technologies (NoDaLiDa/Baltic-HLT 2025)
%D 2025
%8 March
%I University of Tartu Library
%C Tallinn, Estonia
%@ 978-9908-53-109-0
%F gibert-etal-2025-mind
%X We present fast Neural Machine Translation models for 17 diverse languages, developed using Sequence-level Knowledge Distillation. Our selected languages span multiple language families and scripts, including low-resource languages. The distilled models achieve comparable performance while being 10x times faster than transformer-base and 35x times faster than transformer-big architectures. Our experiments reveal that teacher model quality and capacity strongly influence the distillation success, as well as the language script. We also explore the effectiveness of multilingual students. We release publicly our code and models in our Github repository: anonymised.
%U https://aclanthology.org/2025.nodalida-1.21/
%P 209-216
Markdown (Informal)
[Mind the Gap: Diverse NMT Models for Resource-Constrained Environments](https://aclanthology.org/2025.nodalida-1.21/) (Gibert et al., NoDaLiDa 2025)
ACL
- Ona de Gibert, Dayyán O’Brien, Dušan Variš, and Jörg Tiedemann. 2025. Mind the Gap: Diverse NMT Models for Resource-Constrained Environments. In Proceedings of the Joint 25th Nordic Conference on Computational Linguistics and 11th Baltic Conference on Human Language Technologies (NoDaLiDa/Baltic-HLT 2025), pages 209–216, Tallinn, Estonia. University of Tartu Library.