<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="van-nooten-kosar-2026-push">
    <titleInfo>
        <title>Push and Pull: Training Sentence Encoders with Contrastive Losses for Distance-Based Multi-Label Text Classification</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Jens</namePart>
        <namePart type="family">Van Nooten</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Andriy</namePart>
        <namePart type="family">Kosar</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>Despite the potential of Distance-Based Classification (DBC), a method that assigns labels to text by measuring semantic similarity between the text and the label representations, it has received very little attention for Multi-Label Text Classification (MLTC). Previous studies have focused on determining optimal thresholds, reaching promising results with contextual sentence encoders. We demonstrate that the performance of these models can be further improved by training them with contrastive losses, i.e., by bringing text representations closer to the corresponding true label representations in an embedding space. Using three supervised contrastive losses and three sentence encoders (Stella, GIST-Large, and BGE), we evaluated our approach on five English datasets (SemEval, BioTech, Reuters, AAPD, and LitCovid) and one Dutch dataset (EventDNA). The results show consistent substantial improvements over base sentence encoders, thereby narrowing the gap between DBC methods and fine-tuned or zero-shot approaches.</abstract>
    <identifier type="citekey">van-nooten-kosar-2026-push</identifier>
    <identifier type="doi">10.63317/4fiaod5mcdsr</identifier>
    <location>
        <url>https://aclanthology.org/2026.lrec-1.583/</url>
    </location>
    <part>
        <date>2026-05</date>
        <extent unit="page">
            <start>7359</start>
            <end>7379</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Push and Pull: Training Sentence Encoders with Contrastive Losses for Distance-Based Multi-Label Text Classification
%A Van Nooten, Jens
%A Kosar, Andriy
%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 van-nooten-kosar-2026-push
%X Despite the potential of Distance-Based Classification (DBC), a method that assigns labels to text by measuring semantic similarity between the text and the label representations, it has received very little attention for Multi-Label Text Classification (MLTC). Previous studies have focused on determining optimal thresholds, reaching promising results with contextual sentence encoders. We demonstrate that the performance of these models can be further improved by training them with contrastive losses, i.e., by bringing text representations closer to the corresponding true label representations in an embedding space. Using three supervised contrastive losses and three sentence encoders (Stella, GIST-Large, and BGE), we evaluated our approach on five English datasets (SemEval, BioTech, Reuters, AAPD, and LitCovid) and one Dutch dataset (EventDNA). The results show consistent substantial improvements over base sentence encoders, thereby narrowing the gap between DBC methods and fine-tuned or zero-shot approaches.
%R 10.63317/4fiaod5mcdsr
%U https://aclanthology.org/2026.lrec-1.583/
%U https://doi.org/10.63317/4fiaod5mcdsr
%P 7359-7379

Markdown (Informal)

[Push and Pull: Training Sentence Encoders with Contrastive Losses for Distance-Based Multi-Label Text Classification](https://aclanthology.org/2026.lrec-1.583/) (Van Nooten & Kosar, LREC 2026)

• Push and Pull: Training Sentence Encoders with Contrastive Losses for Distance-Based Multi-Label Text Classification (Van Nooten & Kosar, LREC 2026)

ACL

• Jens Van Nooten and Andriy Kosar. 2026. Push and Pull: Training Sentence Encoders with Contrastive Losses for Distance-Based Multi-Label Text Classification. In Proceedings of the Fifteenth Language Resources and Evaluation Conference, pages 7359–7379, Palma de Mallorca, Spain. ELRA Language Resource Association.