@inproceedings{dadason-steingrimsson-2026-optimized,
title = "Optimized for {AI}: Curating the {I}celandic {G}igaword Corpus for Stable {LLM} Training",
author = "Da{\dh}ason, J{\'o}n Fri{\dh}rik and
Steingr{\'i}msson, Stein{\th}{\'o}r",
editor = "Ba{\'n}ski, Piotr and
Knight, Dawn and
Kupietz, Marc and
Witt, Andreas and
Wr{\'o}blewska, Alina",
booktitle = "Proceedings of the 12th Workshop on Challenges in the Management of Large Corpora",
month = may,
year = "2026",
address = "Palma, Mallorca (Spain)",
publisher = "ELRA Language Resources Association (ELRA)",
url = "https://aclanthology.org/2026.cmlc-1.6/",
doi = "10.63317/3uatbht8mdrf",
pages = "49--56",
abstract = "The Icelandic Gigaword Corpus (IGC) is a primary resource for Icelandic NLP, with its current version containing 2.7 billion words of curated text. The IGC is traditionally distributed in a TEI-XML format, a hierarchical structure that allows for rich linguistic annotation and metadata. However, this format introduces significant friction for modern machine learning workflows. Even high-quality curated corpora have been found to contain ``unwanted'' text sequences {--} such as fragmented lists or repetitive boilerplate that may trigger instabilities during training of large language models. In this paper, we present a new processing pipeline designed to optimize the IGC for AI development. We describe a filtering approach focusing on training stability, including fuzzy deduplication to reduce the risk of data leakage, with the aim to provide high-quality data for stable model convergence. Furthermore, we introduce a new JSONL distribution format that bridges the gap between TEI-XML and machine-actionable data, facilitating easier access and safer training for models aiming to work with Icelandic."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="dadason-steingrimsson-2026-optimized">
<titleInfo>
<title>Optimized for AI: Curating the Icelandic Gigaword Corpus for Stable LLM Training</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jón</namePart>
<namePart type="given">Fri\dhrik</namePart>
<namePart type="family">Da\dhason</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Stein\thór</namePart>
<namePart type="family">Steingrímsson</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 12th Workshop on Challenges in the Management of Large Corpora</title>
</titleInfo>
<name type="personal">
<namePart type="given">Piotr</namePart>
<namePart type="family">Bański</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dawn</namePart>
<namePart type="family">Knight</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marc</namePart>
<namePart type="family">Kupietz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andreas</namePart>
<namePart type="family">Witt</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alina</namePart>
<namePart type="family">Wróblewska</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>ELRA Language Resources Association (ELRA)</publisher>
<place>
<placeTerm type="text">Palma, Mallorca (Spain)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>The Icelandic Gigaword Corpus (IGC) is a primary resource for Icelandic NLP, with its current version containing 2.7 billion words of curated text. The IGC is traditionally distributed in a TEI-XML format, a hierarchical structure that allows for rich linguistic annotation and metadata. However, this format introduces significant friction for modern machine learning workflows. Even high-quality curated corpora have been found to contain “unwanted” text sequences – such as fragmented lists or repetitive boilerplate that may trigger instabilities during training of large language models. In this paper, we present a new processing pipeline designed to optimize the IGC for AI development. We describe a filtering approach focusing on training stability, including fuzzy deduplication to reduce the risk of data leakage, with the aim to provide high-quality data for stable model convergence. Furthermore, we introduce a new JSONL distribution format that bridges the gap between TEI-XML and machine-actionable data, facilitating easier access and safer training for models aiming to work with Icelandic.</abstract>
<identifier type="citekey">dadason-steingrimsson-2026-optimized</identifier>
<identifier type="doi">10.63317/3uatbht8mdrf</identifier>
<location>
<url>https://aclanthology.org/2026.cmlc-1.6/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>49</start>
<end>56</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Optimized for AI: Curating the Icelandic Gigaword Corpus for Stable LLM Training
%A Da\dhason, Jón Fri\dhrik
%A Steingrímsson, Stein\thór
%Y Bański, Piotr
%Y Knight, Dawn
%Y Kupietz, Marc
%Y Witt, Andreas
%Y Wróblewska, Alina
%S Proceedings of the 12th Workshop on Challenges in the Management of Large Corpora
%D 2026
%8 May
%I ELRA Language Resources Association (ELRA)
%C Palma, Mallorca (Spain)
%F dadason-steingrimsson-2026-optimized
%X The Icelandic Gigaword Corpus (IGC) is a primary resource for Icelandic NLP, with its current version containing 2.7 billion words of curated text. The IGC is traditionally distributed in a TEI-XML format, a hierarchical structure that allows for rich linguistic annotation and metadata. However, this format introduces significant friction for modern machine learning workflows. Even high-quality curated corpora have been found to contain “unwanted” text sequences – such as fragmented lists or repetitive boilerplate that may trigger instabilities during training of large language models. In this paper, we present a new processing pipeline designed to optimize the IGC for AI development. We describe a filtering approach focusing on training stability, including fuzzy deduplication to reduce the risk of data leakage, with the aim to provide high-quality data for stable model convergence. Furthermore, we introduce a new JSONL distribution format that bridges the gap between TEI-XML and machine-actionable data, facilitating easier access and safer training for models aiming to work with Icelandic.
%R 10.63317/3uatbht8mdrf
%U https://aclanthology.org/2026.cmlc-1.6/
%U https://doi.org/10.63317/3uatbht8mdrf
%P 49-56
Markdown (Informal)
[Optimized for AI: Curating the Icelandic Gigaword Corpus for Stable LLM Training](https://aclanthology.org/2026.cmlc-1.6/) (Daðason & Steingrímsson, CMLC 2026)
ACL