<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="tripodi-2026-representation">
    <titleInfo>
        <title>Representation Learning Enables Language Identification in Zero - and Few -Shot Settings</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Rocco</namePart>
        <namePart type="family">Tripodi</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <genre authority="bibutilsgt">journal article</genre>
    <relatedItem type="host">
        <titleInfo>
            <title>Transactions of the Association for Computational Linguistics</title>
        </titleInfo>
        <originInfo>
            <issuance>continuing</issuance>
            <publisher>MIT Press</publisher>
            <place>
                <placeTerm type="text">Cambridge, MA</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">periodical</genre>
        <genre authority="bibutilsgt">academic journal</genre>
    </relatedItem>
    <abstract>Language Identification (LID) is a key component of multilingual NLP pipelines and multilingual corpus construction. LID errors can create spurious text–language associations, harming downstream model quality, especially for low-resource languages and fine-grained taxonomies involving closely related languages and dialects. We propose LIBE, a representation-based approach that embeds input texts and natural-language descriptions into a shared semantic space with a bi-encoder, enabling LID via similarity search. This formulation supports data-efficient learning and the addition of new languages through their descriptions. To improve discrimination in ambiguous or low-resource cases, we also introduce an optional cross-encoder re-ranking stage. Across multiple benchmarks, LIBE achieves competitive performance with strong LID systems when initialized from pretrained multilingual encoders, while requiring substantially less task-specific LID supervision in several adaptation settings. The hybrid bi-encoder– cross-encoder formulation further improves robustness in challenging open-set scenarios, achieving up to 90% F1 with as few as five labeled examples. Code and data are available at https://github.com/cafoscari-nlp/LIBE.</abstract>
    <identifier type="citekey">tripodi-2026-representation</identifier>
    <identifier type="doi">10.1162/tacl.a.815</identifier>
    <location>
        <url>https://aclanthology.org/2026.tacl-1.118/</url>
    </location>
    <part>
        <date>2026</date>
        <detail type="volume"><number>14</number></detail>
        <extent unit="page">
            <start>2555</start>
            <end>2573</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Journal Article
%T Representation Learning Enables Language Identification in Zero - and Few -Shot Settings
%A Tripodi, Rocco
%J Transactions of the Association for Computational Linguistics
%D 2026
%V 14
%I MIT Press
%C Cambridge, MA
%F tripodi-2026-representation
%X Language Identification (LID) is a key component of multilingual NLP pipelines and multilingual corpus construction. LID errors can create spurious text–language associations, harming downstream model quality, especially for low-resource languages and fine-grained taxonomies involving closely related languages and dialects. We propose LIBE, a representation-based approach that embeds input texts and natural-language descriptions into a shared semantic space with a bi-encoder, enabling LID via similarity search. This formulation supports data-efficient learning and the addition of new languages through their descriptions. To improve discrimination in ambiguous or low-resource cases, we also introduce an optional cross-encoder re-ranking stage. Across multiple benchmarks, LIBE achieves competitive performance with strong LID systems when initialized from pretrained multilingual encoders, while requiring substantially less task-specific LID supervision in several adaptation settings. The hybrid bi-encoder– cross-encoder formulation further improves robustness in challenging open-set scenarios, achieving up to 90% F1 with as few as five labeled examples. Code and data are available at https://github.com/cafoscari-nlp/LIBE.
%R 10.1162/tacl.a.815
%U https://aclanthology.org/2026.tacl-1.118/
%U https://doi.org/10.1162/tacl.a.815
%P 2555-2573

Markdown (Informal)

[Representation Learning Enables Language Identification in Zero - and Few -Shot Settings](https://aclanthology.org/2026.tacl-1.118/) (Tripodi, TACL 2026)

• Representation Learning Enables Language Identification in Zero - and Few -Shot Settings (Tripodi, TACL 2026)

ACL

• Rocco Tripodi. 2026. Representation Learning Enables Language Identification in Zero - and Few -Shot Settings. Transactions of the Association for Computational Linguistics, 14:2555–2573.