<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="anchieta-etal-2026-token">
    <titleInfo>
        <title>Token-Level Pun Location Using Multi-Layer BERT with Mixture of Experts</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Rafael</namePart>
        <namePart type="given">Torres</namePart>
        <namePart type="family">Anchiêta</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Roney</namePart>
        <namePart type="given">Lira</namePart>
        <namePart type="given">de</namePart>
        <namePart type="given">Sales</namePart>
        <namePart type="family">Santos</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Raimundo</namePart>
        <namePart type="given">Santos</namePart>
        <namePart type="family">Moura</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-04</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 17th International Conference on Computational Processing of Portuguese (PROPOR 2026) - Vol. 1</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Marlo</namePart>
            <namePart type="family">Souza</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Iria</namePart>
            <namePart type="family">de-Dios-Flores</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Diana</namePart>
            <namePart type="family">Santos</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Larissa</namePart>
            <namePart type="family">Freitas</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jackson</namePart>
            <namePart type="given">Wilke</namePart>
            <namePart type="given">da</namePart>
            <namePart type="given">Cruz</namePart>
            <namePart type="family">Souza</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Eugénio</namePart>
            <namePart type="family">Ribeiro</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Salvador, Brazil</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-387-6</identifier>
    </relatedItem>
    <abstract>Humor processing remains a complex challenge in Natural Language Processing, particularly the task of pun location, which involves identifying the specific “pivot word” that creates linguistic ambiguity. This paper presents a novel two-stage approach for token-level pun location in Portuguese, addressing the scarcity of research in this language. The first stage uses an ensemble of traditional classifiers to filter out non-pun sentences, thereby reducing class imbalance. The second stage employs a pre-trained BERT encoder combined with a Mixture-of-Experts (MoE) layer to capture specialized linguistic features for token classification. We validate our approach on the Puntuguese corpus, achieving an F-score of 0.74 without requiring post-processing heuristics. Interpretability analyses demonstrate that the MoE experts learn to specialize in distinct mechanisms, such as punchline detection and morphological patterns, thereby confirming the model’s capacity to capture the nuances of humor.</abstract>
    <identifier type="citekey">anchieta-etal-2026-token</identifier>
    <location>
        <url>https://aclanthology.org/2026.propor-1.26/</url>
    </location>
    <part>
        <date>2026-04</date>
        <extent unit="page">
            <start>260</start>
            <end>269</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Token-Level Pun Location Using Multi-Layer BERT with Mixture of Experts
%A Anchiêta, Rafael Torres
%A Santos, Roney Lira de Sales
%A Moura, Raimundo Santos
%Y Souza, Marlo
%Y de-Dios-Flores, Iria
%Y Santos, Diana
%Y Freitas, Larissa
%Y Souza, Jackson Wilke da Cruz
%Y Ribeiro, Eugénio
%S Proceedings of the 17th International Conference on Computational Processing of Portuguese (PROPOR 2026) - Vol. 1
%D 2026
%8 April
%I Association for Computational Linguistics
%C Salvador, Brazil
%@ 979-8-89176-387-6
%F anchieta-etal-2026-token
%X Humor processing remains a complex challenge in Natural Language Processing, particularly the task of pun location, which involves identifying the specific “pivot word” that creates linguistic ambiguity. This paper presents a novel two-stage approach for token-level pun location in Portuguese, addressing the scarcity of research in this language. The first stage uses an ensemble of traditional classifiers to filter out non-pun sentences, thereby reducing class imbalance. The second stage employs a pre-trained BERT encoder combined with a Mixture-of-Experts (MoE) layer to capture specialized linguistic features for token classification. We validate our approach on the Puntuguese corpus, achieving an F-score of 0.74 without requiring post-processing heuristics. Interpretability analyses demonstrate that the MoE experts learn to specialize in distinct mechanisms, such as punchline detection and morphological patterns, thereby confirming the model’s capacity to capture the nuances of humor.
%U https://aclanthology.org/2026.propor-1.26/
%P 260-269

Markdown (Informal)

[Token-Level Pun Location Using Multi-Layer BERT with Mixture of Experts](https://aclanthology.org/2026.propor-1.26/) (Anchiêta et al., PROPOR 2026)

• Token-Level Pun Location Using Multi-Layer BERT with Mixture of Experts (Anchiêta et al., PROPOR 2026)

ACL

• Rafael Torres Anchiêta, Roney Lira de Sales Santos, and Raimundo Santos Moura. 2026. Token-Level Pun Location Using Multi-Layer BERT with Mixture of Experts. In Proceedings of the 17th International Conference on Computational Processing of Portuguese (PROPOR 2026) - Vol. 1, pages 260–269, Salvador, Brazil. Association for Computational Linguistics.