@inproceedings{dumon-etal-2026-beyond,
title = "Beyond Slurs: A Multilingual Analysis of Misogyny Detection Across {D}utch, {E}nglish, and {S}panish",
author = "Dumon, Jirky and
Maladry, Aaron and
Lefever, Els",
editor = "Calabrese, Agostina and
Davidson, Thomas and
de Kock, Christine and
Khurana, Urja and
Marchiori Manerba, Marta and
Piot, Paloma and
Talat, Zeerak",
booktitle = "Proceedings of the 10th Workshop on Online Abuse and Harms ({WOAH})",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.woah-1.23/",
pages = "261--272",
ISBN = "979-8-89176-487-3",
abstract = "The growing volume of gender-based abusive content on social media platforms has increased the need for reliable automatic misogyny detection across languages. While multilingual language models achieve strong overall performance, less is known about how well they handle phenomena such as implicit hate, stereotypes, or humour. In this research, we present a multilingual study of misogyny detection in Dutch, English, and Spanish Twitter data using mmBERT. We introduce a newly collected and manually annotated Dutch dataset and re-annotate English and Spanish HatEval samples using a unified annotation framework. Beyond binary misogyny labels, tweets are annotated for implicitness, aggressiveness, stereotypes, humour, and target type, enabling a fine-grained analysis of model behaviour. Experimental results reveal substantial differences across languages, and show that overall classification performance masks important category-specific weaknesses. While aggressive content is generally detected reliably, implicit misogyny, stereotypes, and humour remain challenging across all three languages. Qualitative analysis further demonstrates that the model over-relies on lexical cues, and struggles with reported speech and stereotype-based misogyny expressed without overtly offensive language. These findings highlight the value of fine-grained multilingual analyses for understanding the limitations of contemporary misogyny-detection systems."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="dumon-etal-2026-beyond">
<titleInfo>
<title>Beyond Slurs: A Multilingual Analysis of Misogyny Detection Across Dutch, English, and Spanish</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jirky</namePart>
<namePart type="family">Dumon</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aaron</namePart>
<namePart type="family">Maladry</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Els</namePart>
<namePart type="family">Lefever</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 10th Workshop on Online Abuse and Harms (WOAH)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Agostina</namePart>
<namePart type="family">Calabrese</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Thomas</namePart>
<namePart type="family">Davidson</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christine</namePart>
<namePart type="family">de Kock</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Urja</namePart>
<namePart type="family">Khurana</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marta</namePart>
<namePart type="family">Marchiori Manerba</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Paloma</namePart>
<namePart type="family">Piot</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zeerak</namePart>
<namePart type="family">Talat</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">979-8-89176-487-3</identifier>
</relatedItem>
<abstract>The growing volume of gender-based abusive content on social media platforms has increased the need for reliable automatic misogyny detection across languages. While multilingual language models achieve strong overall performance, less is known about how well they handle phenomena such as implicit hate, stereotypes, or humour. In this research, we present a multilingual study of misogyny detection in Dutch, English, and Spanish Twitter data using mmBERT. We introduce a newly collected and manually annotated Dutch dataset and re-annotate English and Spanish HatEval samples using a unified annotation framework. Beyond binary misogyny labels, tweets are annotated for implicitness, aggressiveness, stereotypes, humour, and target type, enabling a fine-grained analysis of model behaviour. Experimental results reveal substantial differences across languages, and show that overall classification performance masks important category-specific weaknesses. While aggressive content is generally detected reliably, implicit misogyny, stereotypes, and humour remain challenging across all three languages. Qualitative analysis further demonstrates that the model over-relies on lexical cues, and struggles with reported speech and stereotype-based misogyny expressed without overtly offensive language. These findings highlight the value of fine-grained multilingual analyses for understanding the limitations of contemporary misogyny-detection systems.</abstract>
<identifier type="citekey">dumon-etal-2026-beyond</identifier>
<location>
<url>https://aclanthology.org/2026.woah-1.23/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>261</start>
<end>272</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Beyond Slurs: A Multilingual Analysis of Misogyny Detection Across Dutch, English, and Spanish
%A Dumon, Jirky
%A Maladry, Aaron
%A Lefever, Els
%Y Calabrese, Agostina
%Y Davidson, Thomas
%Y de Kock, Christine
%Y Khurana, Urja
%Y Marchiori Manerba, Marta
%Y Piot, Paloma
%Y Talat, Zeerak
%S Proceedings of the 10th Workshop on Online Abuse and Harms (WOAH)
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-487-3
%F dumon-etal-2026-beyond
%X The growing volume of gender-based abusive content on social media platforms has increased the need for reliable automatic misogyny detection across languages. While multilingual language models achieve strong overall performance, less is known about how well they handle phenomena such as implicit hate, stereotypes, or humour. In this research, we present a multilingual study of misogyny detection in Dutch, English, and Spanish Twitter data using mmBERT. We introduce a newly collected and manually annotated Dutch dataset and re-annotate English and Spanish HatEval samples using a unified annotation framework. Beyond binary misogyny labels, tweets are annotated for implicitness, aggressiveness, stereotypes, humour, and target type, enabling a fine-grained analysis of model behaviour. Experimental results reveal substantial differences across languages, and show that overall classification performance masks important category-specific weaknesses. While aggressive content is generally detected reliably, implicit misogyny, stereotypes, and humour remain challenging across all three languages. Qualitative analysis further demonstrates that the model over-relies on lexical cues, and struggles with reported speech and stereotype-based misogyny expressed without overtly offensive language. These findings highlight the value of fine-grained multilingual analyses for understanding the limitations of contemporary misogyny-detection systems.
%U https://aclanthology.org/2026.woah-1.23/
%P 261-272
Markdown (Informal)
[Beyond Slurs: A Multilingual Analysis of Misogyny Detection Across Dutch, English, and Spanish](https://aclanthology.org/2026.woah-1.23/) (Dumon et al., WOAH 2026)
ACL