<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kurtyigit-etal-2026-learning">
    <titleInfo>
        <title>Learning the Cue or Learning the Word? Analyzing Generalization in Metaphor Detection for Verbs</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Sinan</namePart>
        <namePart type="family">Kurtyigit</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Sabine</namePart>
        <namePart type="family">Schulte im Walde</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Alexander</namePart>
        <namePart type="family">Fraser</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 2026 Conference on Empirical Methods in Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Hung-Yi</namePart>
            <namePart type="family">Lee</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sunipa</namePart>
            <namePart type="family">Dev</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jessy</namePart>
            <namePart type="family">Li</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Desmond</namePart>
            <namePart type="family">Elliott</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">979-8-89176-459-0</identifier>
    </relatedItem>
    <abstract>We analyze the generalization capabilities of a RoBERTa-based metaphor detection model on English verbs using the VU Amsterdam Metaphor Corpus. Since RoBERTa encoders are widely-used in metaphor detection, determining whether they rely on lexical memorization or learn transferable patterns is critical. We introduce a controlled lexical hold-out setup where all instances of selected target lemmas are strictly excluded from fine-tuning. We then compare model predictions on these Held-out lemmas against Exposed lemmas (verbs encountered during fine-tuning). While the model performs best on Exposed lemmas, it maintains robust performance on Held-out lemmas. This success does not correlate with verb lemma frequency, suggesting the generalization is likely not driven by frequency bias in pre-training. Further analysis reveals that sentence context alone is sufficient to match full-model performance on Held-out lemmas, whereas static verb-level embeddings are not. Geometrically, we find that the encoder projects Held-out instances into the same task-relevant subspaces as Exposed instances. Together, these results suggest that the model exhibits generalization capabilities, which appear to be driven by “learning the cue” (transferable contextual patterns) rather than “learning the word” (verb-specific memorization).</abstract>
    <identifier type="citekey">kurtyigit-etal-2026-learning</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.136/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>3136</start>
            <end>3146</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Learning the Cue or Learning the Word? Analyzing Generalization in Metaphor Detection for Verbs
%A Kurtyigit, Sinan
%A Schulte im Walde, Sabine
%A Fraser, Alexander
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F kurtyigit-etal-2026-learning
%X We analyze the generalization capabilities of a RoBERTa-based metaphor detection model on English verbs using the VU Amsterdam Metaphor Corpus. Since RoBERTa encoders are widely-used in metaphor detection, determining whether they rely on lexical memorization or learn transferable patterns is critical. We introduce a controlled lexical hold-out setup where all instances of selected target lemmas are strictly excluded from fine-tuning. We then compare model predictions on these Held-out lemmas against Exposed lemmas (verbs encountered during fine-tuning). While the model performs best on Exposed lemmas, it maintains robust performance on Held-out lemmas. This success does not correlate with verb lemma frequency, suggesting the generalization is likely not driven by frequency bias in pre-training. Further analysis reveals that sentence context alone is sufficient to match full-model performance on Held-out lemmas, whereas static verb-level embeddings are not. Geometrically, we find that the encoder projects Held-out instances into the same task-relevant subspaces as Exposed instances. Together, these results suggest that the model exhibits generalization capabilities, which appear to be driven by “learning the cue” (transferable contextual patterns) rather than “learning the word” (verb-specific memorization).
%U https://aclanthology.org/2026.emnlp-main.136/
%P 3136-3146

Markdown (Informal)

[Learning the Cue or Learning the Word? Analyzing Generalization in Metaphor Detection for Verbs](https://aclanthology.org/2026.emnlp-main.136/) (Kurtyigit et al., EMNLP 2026)

• Learning the Cue or Learning the Word? Analyzing Generalization in Metaphor Detection for Verbs (Kurtyigit et al., EMNLP 2026)

ACL

• Sinan Kurtyigit, Sabine Schulte im Walde, and Alexander Fraser. 2026. Learning the Cue or Learning the Word? Analyzing Generalization in Metaphor Detection for Verbs. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 3136–3146, Budapest, Hungary. Association for Computational Linguistics.