@inproceedings{ji-etal-2026-data,
title = "Data-Centric Continual Pre-training for 500+ Languages: A New Bilingual Translation Corpus and Multilingual Models",
author = {Ji, Shaoxiong and
Li, Zihao and
Paavola, Jaakko and
Luo, Hengyu and
Tiedemann, J{\"o}rg},
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-acl.937/",
pages = "18776--18807",
ISBN = "979-8-89176-395-1",
abstract = "This paper investigates a critical design decision in the practice of massively multilingual continual pre-training {---} the inclusion of parallel data. Specifically, we study the impact of bilingual translation data for massively multilingual language adaptation of the Llama 3 family of models to 500 languages. To this end, we construct a bilingual translation corpus named OUR{\_}DATA, containing data from more than 2,500 language pairs. Subsequently, we develop the OUR{\_}MODEL Llama 3 suite of four massively multilingual models {---} continually pre-trained from the Llama 3 family of base models extensively on diverse data mixes up to 671B tokens {---} and explore the effect of continual pre-training with or without bilingual translation data. Comprehensive evaluation across 7 tasks and 12 benchmarks demonstrates that bilingual data tends to enhance language transfer and performance, particularly for low-resource languages. We open-source the OUR{\_}DATA corpus, OUR{\_}MODEL Llama 3 suite artefacts, code, and model generations."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ji-etal-2026-data">
<titleInfo>
<title>Data-Centric Continual Pre-training for 500+ Languages: A New Bilingual Translation Corpus and Multilingual Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Shaoxiong</namePart>
<namePart type="family">Ji</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zihao</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jaakko</namePart>
<namePart type="family">Paavola</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hengyu</namePart>
<namePart type="family">Luo</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>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-395-1</identifier>
</relatedItem>
<abstract>This paper investigates a critical design decision in the practice of massively multilingual continual pre-training — the inclusion of parallel data. Specifically, we study the impact of bilingual translation data for massively multilingual language adaptation of the Llama 3 family of models to 500 languages. To this end, we construct a bilingual translation corpus named OUR_DATA, containing data from more than 2,500 language pairs. Subsequently, we develop the OUR_MODEL Llama 3 suite of four massively multilingual models — continually pre-trained from the Llama 3 family of base models extensively on diverse data mixes up to 671B tokens — and explore the effect of continual pre-training with or without bilingual translation data. Comprehensive evaluation across 7 tasks and 12 benchmarks demonstrates that bilingual data tends to enhance language transfer and performance, particularly for low-resource languages. We open-source the OUR_DATA corpus, OUR_MODEL Llama 3 suite artefacts, code, and model generations.</abstract>
<identifier type="citekey">ji-etal-2026-data</identifier>
<location>
<url>https://aclanthology.org/2026.findings-acl.937/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>18776</start>
<end>18807</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Data-Centric Continual Pre-training for 500+ Languages: A New Bilingual Translation Corpus and Multilingual Models
%A Ji, Shaoxiong
%A Li, Zihao
%A Paavola, Jaakko
%A Luo, Hengyu
%A Tiedemann, Jörg
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Findings of the Association for Computational Linguistics: ACL 2026
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-395-1
%F ji-etal-2026-data
%X This paper investigates a critical design decision in the practice of massively multilingual continual pre-training — the inclusion of parallel data. Specifically, we study the impact of bilingual translation data for massively multilingual language adaptation of the Llama 3 family of models to 500 languages. To this end, we construct a bilingual translation corpus named OUR_DATA, containing data from more than 2,500 language pairs. Subsequently, we develop the OUR_MODEL Llama 3 suite of four massively multilingual models — continually pre-trained from the Llama 3 family of base models extensively on diverse data mixes up to 671B tokens — and explore the effect of continual pre-training with or without bilingual translation data. Comprehensive evaluation across 7 tasks and 12 benchmarks demonstrates that bilingual data tends to enhance language transfer and performance, particularly for low-resource languages. We open-source the OUR_DATA corpus, OUR_MODEL Llama 3 suite artefacts, code, and model generations.
%U https://aclanthology.org/2026.findings-acl.937/
%P 18776-18807
Markdown (Informal)
[Data-Centric Continual Pre-training for 500+ Languages: A New Bilingual Translation Corpus and Multilingual Models](https://aclanthology.org/2026.findings-acl.937/) (Ji et al., Findings 2026)
ACL