@inproceedings{roussopoulou-etal-2026-hurtnet,
title = "{H}urt{N}et: a Multilingual Dictionary of Hurtful Words",
author = {Roussopoulou, Maria Alexandra and
Di Palma, Eliana and
Di Fabio, Andrea and
Mahmoud Wizani, Adel and
Stamou, Vivian and
Ollagnier, Ana{\"i}s and
Osenova, Petya and
Basile, Valerio and
Markantonatou, Stella},
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-1.43/",
pages = "507--523",
ISBN = "979-8-89176-487-3",
abstract = "We study the problem of preserving offensive load in LLM-generated translations, focusing on how offensive words retain their load across languages. We draw on HurtNet, a novel language resource consisting of five multilingual (Arabic, French, Bulgarian, Greek, and Italian) dictionaries of hurtful terms, with multiple definitions and examples of usage. We draw on HurtNet to identify offensive senses shared by the 5 languages and the terms that denote them, and feed translations with this knowledge. We evaluate the approach through quantitative and qualitative analysis. A series of knowledge-injection zero-shot translation experiments were carried out with four different prompts that gradually increase injected linguistic knowledge, and three LLMs {---}an open use, a translation-dedicated, and a commercial one. Our results show that translations better preserving offensive load are produced when the prompt explicitly instructs the model about the translated offensive term and that the method is beneficial for those cases where the literal translation of the term does not preserve its offensive load."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="roussopoulou-etal-2026-hurtnet">
<titleInfo>
<title>HurtNet: a Multilingual Dictionary of Hurtful Words</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="given">Alexandra</namePart>
<namePart type="family">Roussopoulou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Eliana</namePart>
<namePart type="family">Di Palma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andrea</namePart>
<namePart type="family">Di Fabio</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Adel</namePart>
<namePart type="family">Mahmoud Wizani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vivian</namePart>
<namePart type="family">Stamou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anaïs</namePart>
<namePart type="family">Ollagnier</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Petya</namePart>
<namePart type="family">Osenova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Valerio</namePart>
<namePart type="family">Basile</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Stella</namePart>
<namePart type="family">Markantonatou</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>We study the problem of preserving offensive load in LLM-generated translations, focusing on how offensive words retain their load across languages. We draw on HurtNet, a novel language resource consisting of five multilingual (Arabic, French, Bulgarian, Greek, and Italian) dictionaries of hurtful terms, with multiple definitions and examples of usage. We draw on HurtNet to identify offensive senses shared by the 5 languages and the terms that denote them, and feed translations with this knowledge. We evaluate the approach through quantitative and qualitative analysis. A series of knowledge-injection zero-shot translation experiments were carried out with four different prompts that gradually increase injected linguistic knowledge, and three LLMs —an open use, a translation-dedicated, and a commercial one. Our results show that translations better preserving offensive load are produced when the prompt explicitly instructs the model about the translated offensive term and that the method is beneficial for those cases where the literal translation of the term does not preserve its offensive load.</abstract>
<identifier type="citekey">roussopoulou-etal-2026-hurtnet</identifier>
<location>
<url>https://aclanthology.org/2026.woah-1.43/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>507</start>
<end>523</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T HurtNet: a Multilingual Dictionary of Hurtful Words
%A Roussopoulou, Maria Alexandra
%A Di Palma, Eliana
%A Di Fabio, Andrea
%A Mahmoud Wizani, Adel
%A Stamou, Vivian
%A Ollagnier, Anaïs
%A Osenova, Petya
%A Basile, Valerio
%A Markantonatou, Stella
%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 roussopoulou-etal-2026-hurtnet
%X We study the problem of preserving offensive load in LLM-generated translations, focusing on how offensive words retain their load across languages. We draw on HurtNet, a novel language resource consisting of five multilingual (Arabic, French, Bulgarian, Greek, and Italian) dictionaries of hurtful terms, with multiple definitions and examples of usage. We draw on HurtNet to identify offensive senses shared by the 5 languages and the terms that denote them, and feed translations with this knowledge. We evaluate the approach through quantitative and qualitative analysis. A series of knowledge-injection zero-shot translation experiments were carried out with four different prompts that gradually increase injected linguistic knowledge, and three LLMs —an open use, a translation-dedicated, and a commercial one. Our results show that translations better preserving offensive load are produced when the prompt explicitly instructs the model about the translated offensive term and that the method is beneficial for those cases where the literal translation of the term does not preserve its offensive load.
%U https://aclanthology.org/2026.woah-1.43/
%P 507-523
Markdown (Informal)
[HurtNet: a Multilingual Dictionary of Hurtful Words](https://aclanthology.org/2026.woah-1.43/) (Roussopoulou et al., WOAH 2026)
ACL
- Maria Alexandra Roussopoulou, Eliana Di Palma, Andrea Di Fabio, Adel Mahmoud Wizani, Vivian Stamou, Anaïs Ollagnier, Petya Osenova, Valerio Basile, and Stella Markantonatou. 2026. HurtNet: a Multilingual Dictionary of Hurtful Words. In Proceedings of the 10th Workshop on Online Abuse and Harms (WOAH), pages 507–523, Budapest, Hungary. Association for Computational Linguistics.