<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="rahbar-2026-curse">
    <titleInfo>
        <title>The Curse of Multilinguality in Lexical Normalization</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Saman</namePart>
        <namePart type="family">Rahbar</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-10</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 11th Workshop on Natural User-generated Text</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">JinYeong</namePart>
            <namePart type="family">Bak</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Rob</namePart>
            <namePart type="family">van der Goot</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Hyeju</namePart>
            <namePart type="family">Jang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Weerayut</namePart>
            <namePart type="family">Buaphet</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Alan</namePart>
            <namePart type="family">Ramponi</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Wei</namePart>
            <namePart type="family">Xu</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Alan</namePart>
            <namePart type="family">Ritter</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Budapest, Hungary</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-470-5</identifier>
    </relatedItem>
    <abstract>Lexical normalization rewrites the noisy, nonstandard words that fill user-generated text (tmrw, u, gr8) into their standard forms. Because labelled data is scarce for most languages, a popular shortcut is to train a single model on many languages at once. We ask a simple question: how many languages should such a model be trained on? Using one fixed-capacity character-level model and twelve languages from a standard benchmark, we vary the number of jointly trained languages from one to twelve and measure per-language accuracy. We find a clear curse of multilinguality: accuracy is highest when a language is trained with only a few others, often just one to four, and then falls steadily and substantially, dropping by about forty percent as the rest are piled on. A control that holds the total amount of training data constant makes the decline arrive sooner and fall further, which points to competition among the languages for one fixed-size model rather than to how much data is available. We also test whether a language’s typological distance from the others predicts its ideal number of cotraining languages, and find no dependable rule: any apparent relationship rests on a couple of languages and does not hold up. For compact normalization models, less can be more: a few languages beat pooling everything into a single model.</abstract>
    <identifier type="citekey">rahbar-2026-curse</identifier>
    <location>
        <url>https://aclanthology.org/2026.wnut-1.22/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>325</start>
            <end>331</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T The Curse of Multilinguality in Lexical Normalization
%A Rahbar, Saman
%Y Bak, JinYeong
%Y van der Goot, Rob
%Y Jang, Hyeju
%Y Buaphet, Weerayut
%Y Ramponi, Alan
%Y Xu, Wei
%Y Ritter, Alan
%S Proceedings of the 11th Workshop on Natural User-generated Text
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-470-5
%F rahbar-2026-curse
%X Lexical normalization rewrites the noisy, nonstandard words that fill user-generated text (tmrw, u, gr8) into their standard forms. Because labelled data is scarce for most languages, a popular shortcut is to train a single model on many languages at once. We ask a simple question: how many languages should such a model be trained on? Using one fixed-capacity character-level model and twelve languages from a standard benchmark, we vary the number of jointly trained languages from one to twelve and measure per-language accuracy. We find a clear curse of multilinguality: accuracy is highest when a language is trained with only a few others, often just one to four, and then falls steadily and substantially, dropping by about forty percent as the rest are piled on. A control that holds the total amount of training data constant makes the decline arrive sooner and fall further, which points to competition among the languages for one fixed-size model rather than to how much data is available. We also test whether a language’s typological distance from the others predicts its ideal number of cotraining languages, and find no dependable rule: any apparent relationship rests on a couple of languages and does not hold up. For compact normalization models, less can be more: a few languages beat pooling everything into a single model.
%U https://aclanthology.org/2026.wnut-1.22/
%P 325-331

Markdown (Informal)

[The Curse of Multilinguality in Lexical Normalization](https://aclanthology.org/2026.wnut-1.22/) (Rahbar, WNUT 2026)

• The Curse of Multilinguality in Lexical Normalization (Rahbar, WNUT 2026)

ACL

• Saman Rahbar. 2026. The Curse of Multilinguality in Lexical Normalization. In Proceedings of the 11th Workshop on Natural User-generated Text, pages 325–331, Budapest, Hungary. Association for Computational Linguistics.