@inproceedings{stenlund-etal-2026-synthetic,
title = "Synthetic Instruction Generation for Low-Resource {N}ordic Languages: Viability and Limitations in {LLM} Instruction-Tuning",
author = "Stenlund, Mathias and
Simonsen, Annika and
Bungum, Lars and
Ebert, Jan and
Wang, Jiangtao and
Filatov, Oleg and
Myneni, Hemanadhan and
Riedel, Morris and
Einarsson, Hafsteinn",
editor = "Piperidis, Stelios and
Bel, N{\'u}ria and
van den Heuvel, Henk and
Ide, Nancy and
Krek, Simon and
Toral, Antonio",
booktitle = "Proceedings of the Fifteenth Language Resources and Evaluation Conference",
month = may,
year = "2026",
address = "Palma de Mallorca, Spain",
publisher = "ELRA Language Resource Association",
url = "https://aclanthology.org/2026.lrec-1.838/",
doi = "10.63317/3e73sy24wup3",
pages = "10688--10698",
abstract = "Pretrained large language models (LLMs) gain instruction-following abilities through instruction-tuning, a method which relies on datasets of instruction{--}response pairs. However, for low-resource languages, collecting human-authored instructions is costly, raising the question of whether synthetic instructions can substitute human-authored instructions for non-English languages. We compare instruction-tuning of a smaller pretrained LLM in four Nordic languages using (a) human-authored instructions paired with synthetic responses and (b) fully synthetic instruction{--}response pairs generated with a minimal-effort pipeline. Native-speaker evaluations show that models instruction-tuned on synthetic instructions perform on par with those trained on human-authored instructions for the largest Nordic languages, suggesting that minimal-effort synthetic instructions can serve as a practical alternative. In contrast, response quality deteriorates sharply for Icelandic, underscoring the limitations of current synthetic data generation pipelines when the LLM competence in the target language is weak. Overall, our results highlight that while synthetic instructions can enable cost-efficient instruction-tuning for the largest Nordic languages, they remain insufficient for Icelandic, clarifying when minimal-effort synthetic approaches suffice and when they fall short."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="stenlund-etal-2026-synthetic">
<titleInfo>
<title>Synthetic Instruction Generation for Low-Resource Nordic Languages: Viability and Limitations in LLM Instruction-Tuning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mathias</namePart>
<namePart type="family">Stenlund</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">Lars</namePart>
<namePart type="family">Bungum</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jan</namePart>
<namePart type="family">Ebert</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiangtao</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Oleg</namePart>
<namePart type="family">Filatov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hemanadhan</namePart>
<namePart type="family">Myneni</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Morris</namePart>
<namePart type="family">Riedel</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>2026-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Fifteenth Language Resources and Evaluation Conference</title>
</titleInfo>
<name type="personal">
<namePart type="given">Stelios</namePart>
<namePart type="family">Piperidis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Núria</namePart>
<namePart type="family">Bel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Henk</namePart>
<namePart type="family">van den Heuvel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nancy</namePart>
<namePart type="family">Ide</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Simon</namePart>
<namePart type="family">Krek</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Antonio</namePart>
<namePart type="family">Toral</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>ELRA Language Resource Association</publisher>
<place>
<placeTerm type="text">Palma de Mallorca, Spain</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Pretrained large language models (LLMs) gain instruction-following abilities through instruction-tuning, a method which relies on datasets of instruction–response pairs. However, for low-resource languages, collecting human-authored instructions is costly, raising the question of whether synthetic instructions can substitute human-authored instructions for non-English languages. We compare instruction-tuning of a smaller pretrained LLM in four Nordic languages using (a) human-authored instructions paired with synthetic responses and (b) fully synthetic instruction–response pairs generated with a minimal-effort pipeline. Native-speaker evaluations show that models instruction-tuned on synthetic instructions perform on par with those trained on human-authored instructions for the largest Nordic languages, suggesting that minimal-effort synthetic instructions can serve as a practical alternative. In contrast, response quality deteriorates sharply for Icelandic, underscoring the limitations of current synthetic data generation pipelines when the LLM competence in the target language is weak. Overall, our results highlight that while synthetic instructions can enable cost-efficient instruction-tuning for the largest Nordic languages, they remain insufficient for Icelandic, clarifying when minimal-effort synthetic approaches suffice and when they fall short.</abstract>
<identifier type="citekey">stenlund-etal-2026-synthetic</identifier>
<identifier type="doi">10.63317/3e73sy24wup3</identifier>
<location>
<url>https://aclanthology.org/2026.lrec-1.838/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>10688</start>
<end>10698</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Synthetic Instruction Generation for Low-Resource Nordic Languages: Viability and Limitations in LLM Instruction-Tuning
%A Stenlund, Mathias
%A Simonsen, Annika
%A Bungum, Lars
%A Ebert, Jan
%A Wang, Jiangtao
%A Filatov, Oleg
%A Myneni, Hemanadhan
%A Riedel, Morris
%A Einarsson, Hafsteinn
%Y Piperidis, Stelios
%Y Bel, Núria
%Y van den Heuvel, Henk
%Y Ide, Nancy
%Y Krek, Simon
%Y Toral, Antonio
%S Proceedings of the Fifteenth Language Resources and Evaluation Conference
%D 2026
%8 May
%I ELRA Language Resource Association
%C Palma de Mallorca, Spain
%F stenlund-etal-2026-synthetic
%X Pretrained large language models (LLMs) gain instruction-following abilities through instruction-tuning, a method which relies on datasets of instruction–response pairs. However, for low-resource languages, collecting human-authored instructions is costly, raising the question of whether synthetic instructions can substitute human-authored instructions for non-English languages. We compare instruction-tuning of a smaller pretrained LLM in four Nordic languages using (a) human-authored instructions paired with synthetic responses and (b) fully synthetic instruction–response pairs generated with a minimal-effort pipeline. Native-speaker evaluations show that models instruction-tuned on synthetic instructions perform on par with those trained on human-authored instructions for the largest Nordic languages, suggesting that minimal-effort synthetic instructions can serve as a practical alternative. In contrast, response quality deteriorates sharply for Icelandic, underscoring the limitations of current synthetic data generation pipelines when the LLM competence in the target language is weak. Overall, our results highlight that while synthetic instructions can enable cost-efficient instruction-tuning for the largest Nordic languages, they remain insufficient for Icelandic, clarifying when minimal-effort synthetic approaches suffice and when they fall short.
%R 10.63317/3e73sy24wup3
%U https://aclanthology.org/2026.lrec-1.838/
%U https://doi.org/10.63317/3e73sy24wup3
%P 10688-10698
Markdown (Informal)
[Synthetic Instruction Generation for Low-Resource Nordic Languages: Viability and Limitations in LLM Instruction-Tuning](https://aclanthology.org/2026.lrec-1.838/) (Stenlund et al., LREC 2026)
ACL
- Mathias Stenlund, Annika Simonsen, Lars Bungum, Jan Ebert, Jiangtao Wang, Oleg Filatov, Hemanadhan Myneni, Morris Riedel, and Hafsteinn Einarsson. 2026. Synthetic Instruction Generation for Low-Resource Nordic Languages: Viability and Limitations in LLM Instruction-Tuning. In Proceedings of the Fifteenth Language Resources and Evaluation Conference, pages 10688–10698, Palma de Mallorca, Spain. ELRA Language Resource Association.