@inproceedings{scalvini-etal-2025-rethinking,
title = "Rethinking Low-Resource {MT:} The Surprising Effectiveness of Fine-Tuned Multilingual Models in the {LLM} Age",
author = "Scalvini, Barbara and
Debess, Iben Nyholm and
Simonsen, Annika and
Einarsson, Hafsteinn",
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-1.62/",
pages = "609--621",
ISBN = "978-9908-53-109-0",
abstract = "This study challenges the current paradigm shift in machine translation, where large language models (LLMs) are gaining prominence over traditional neural machine translation models, with a focus on English-to-Faroese translation. We compare the performance of various models, including fine-tuned multilingual models, LLMs (GPT-SW3, Llama 3.1), and closed-source models (Claude 3.5, GPT-4). Our findings show that a fine-tuned NLLB model outperforms most LLMs, including some larger models, in both automatic and human evaluations. We also demonstrate the effectiveness of using LLM-generated synthetic data for fine-tuning. While closed-source models like Claude 3.5 perform best overall, the competitive performance of smaller, fine-tuned models suggests a more nuanced approach to low-resource machine translation. Our results highlight the potential of specialized multilingual models and the importance of language-specific knowledge. We discuss implications for resource allocation in low-resource settings and suggest future directions for improving low-resource machine translation, including targeted data creation and more comprehensive evaluation methodologies."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="scalvini-etal-2025-rethinking">
<titleInfo>
<title>Rethinking Low-Resource MT: The Surprising Effectiveness of Fine-Tuned Multilingual Models in the LLM Age</title>
</titleInfo>
<name type="personal">
<namePart type="given">Barbara</namePart>
<namePart type="family">Scalvini</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Iben</namePart>
<namePart type="given">Nyholm</namePart>
<namePart type="family">Debess</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Annika</namePart>
<namePart type="family">Simonsen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hafsteinn</namePart>
<namePart type="family">Einarsson</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>This study challenges the current paradigm shift in machine translation, where large language models (LLMs) are gaining prominence over traditional neural machine translation models, with a focus on English-to-Faroese translation. We compare the performance of various models, including fine-tuned multilingual models, LLMs (GPT-SW3, Llama 3.1), and closed-source models (Claude 3.5, GPT-4). Our findings show that a fine-tuned NLLB model outperforms most LLMs, including some larger models, in both automatic and human evaluations. We also demonstrate the effectiveness of using LLM-generated synthetic data for fine-tuning. While closed-source models like Claude 3.5 perform best overall, the competitive performance of smaller, fine-tuned models suggests a more nuanced approach to low-resource machine translation. Our results highlight the potential of specialized multilingual models and the importance of language-specific knowledge. We discuss implications for resource allocation in low-resource settings and suggest future directions for improving low-resource machine translation, including targeted data creation and more comprehensive evaluation methodologies.</abstract>
<identifier type="citekey">scalvini-etal-2025-rethinking</identifier>
<location>
<url>https://aclanthology.org/2025.nodalida-1.62/</url>
</location>
<part>
<date>2025-03</date>
<extent unit="page">
<start>609</start>
<end>621</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Rethinking Low-Resource MT: The Surprising Effectiveness of Fine-Tuned Multilingual Models in the LLM Age
%A Scalvini, Barbara
%A Debess, Iben Nyholm
%A Simonsen, Annika
%A Einarsson, Hafsteinn
%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 scalvini-etal-2025-rethinking
%X This study challenges the current paradigm shift in machine translation, where large language models (LLMs) are gaining prominence over traditional neural machine translation models, with a focus on English-to-Faroese translation. We compare the performance of various models, including fine-tuned multilingual models, LLMs (GPT-SW3, Llama 3.1), and closed-source models (Claude 3.5, GPT-4). Our findings show that a fine-tuned NLLB model outperforms most LLMs, including some larger models, in both automatic and human evaluations. We also demonstrate the effectiveness of using LLM-generated synthetic data for fine-tuning. While closed-source models like Claude 3.5 perform best overall, the competitive performance of smaller, fine-tuned models suggests a more nuanced approach to low-resource machine translation. Our results highlight the potential of specialized multilingual models and the importance of language-specific knowledge. We discuss implications for resource allocation in low-resource settings and suggest future directions for improving low-resource machine translation, including targeted data creation and more comprehensive evaluation methodologies.
%U https://aclanthology.org/2025.nodalida-1.62/
%P 609-621
Markdown (Informal)
[Rethinking Low-Resource MT: The Surprising Effectiveness of Fine-Tuned Multilingual Models in the LLM Age](https://aclanthology.org/2025.nodalida-1.62/) (Scalvini et al., NoDaLiDa 2025)
ACL