@inproceedings{shabayek-etal-2026-exploring,
title = "Exploring {LLM} Performance for sexist hate speech detection on {F}rench social media",
author = "Shabayek, Shaden and
Penzo, Camilla and
Lee, Eunkyung and
Benamara, Farah and
Moriceau, V{\'e}ronique and
Dashyan, Gohar",
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-1.21/",
pages = "234--248",
ISBN = "979-8-89176-487-3",
abstract = "Deploying Large Language Models (LLMs) for content moderation on social media platforms introduces new challenges related to potential errors and biases, especially for speech-related policies. In this paper, we evaluate a diverse sample of 20 LLMs in zero-shot classification, for the specific case of sexist hate speech in French; a challenging task due to the scarcity of corpora in French and the English-centric research. We elaborate multiple protocols to explore the models' understanding of the task and sensitivity to the language of the prompts and class labels. We show that small or medium size models can outperform much larger models, supporting the paradigm that scale is not the main driver of performance. We also demonstrate that models' performance is not significantly altered when receiving specific guidelines about how to annotate potentially sexist content in French. Qualitative error analysis reveals that guidelines can help capture edge-cases (e.g. condemn sexism), but those are off-set by new misclassifications. This highlights a limitation of zero-shot prompting for content moderation, where specific instructions are crucial for adapting to different languages and cultures."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="shabayek-etal-2026-exploring">
<titleInfo>
<title>Exploring LLM Performance for sexist hate speech detection on French social media</title>
</titleInfo>
<name type="personal">
<namePart type="given">Shaden</namePart>
<namePart type="family">Shabayek</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Camilla</namePart>
<namePart type="family">Penzo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Eunkyung</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Farah</namePart>
<namePart type="family">Benamara</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Véronique</namePart>
<namePart type="family">Moriceau</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gohar</namePart>
<namePart type="family">Dashyan</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>Deploying Large Language Models (LLMs) for content moderation on social media platforms introduces new challenges related to potential errors and biases, especially for speech-related policies. In this paper, we evaluate a diverse sample of 20 LLMs in zero-shot classification, for the specific case of sexist hate speech in French; a challenging task due to the scarcity of corpora in French and the English-centric research. We elaborate multiple protocols to explore the models’ understanding of the task and sensitivity to the language of the prompts and class labels. We show that small or medium size models can outperform much larger models, supporting the paradigm that scale is not the main driver of performance. We also demonstrate that models’ performance is not significantly altered when receiving specific guidelines about how to annotate potentially sexist content in French. Qualitative error analysis reveals that guidelines can help capture edge-cases (e.g. condemn sexism), but those are off-set by new misclassifications. This highlights a limitation of zero-shot prompting for content moderation, where specific instructions are crucial for adapting to different languages and cultures.</abstract>
<identifier type="citekey">shabayek-etal-2026-exploring</identifier>
<location>
<url>https://aclanthology.org/2026.woah-1.21/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>234</start>
<end>248</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Exploring LLM Performance for sexist hate speech detection on French social media
%A Shabayek, Shaden
%A Penzo, Camilla
%A Lee, Eunkyung
%A Benamara, Farah
%A Moriceau, Véronique
%A Dashyan, Gohar
%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 shabayek-etal-2026-exploring
%X Deploying Large Language Models (LLMs) for content moderation on social media platforms introduces new challenges related to potential errors and biases, especially for speech-related policies. In this paper, we evaluate a diverse sample of 20 LLMs in zero-shot classification, for the specific case of sexist hate speech in French; a challenging task due to the scarcity of corpora in French and the English-centric research. We elaborate multiple protocols to explore the models’ understanding of the task and sensitivity to the language of the prompts and class labels. We show that small or medium size models can outperform much larger models, supporting the paradigm that scale is not the main driver of performance. We also demonstrate that models’ performance is not significantly altered when receiving specific guidelines about how to annotate potentially sexist content in French. Qualitative error analysis reveals that guidelines can help capture edge-cases (e.g. condemn sexism), but those are off-set by new misclassifications. This highlights a limitation of zero-shot prompting for content moderation, where specific instructions are crucial for adapting to different languages and cultures.
%U https://aclanthology.org/2026.woah-1.21/
%P 234-248
Markdown (Informal)
[Exploring LLM Performance for sexist hate speech detection on French social media](https://aclanthology.org/2026.woah-1.21/) (Shabayek et al., WOAH 2026)
ACL