@inproceedings{russo-2025-trenteam,
title = "{T}ren{T}eam at Multilingual Counterspeech Generation: Multilingual Passage Re-Ranking Approaches for Knowledge-Driven Counterspeech Generation Against Hate",
author = "Russo, Daniel",
editor = "Bonaldi, Helena and
Vallecillo-Rodr{\'i}guez, Mar{\'i}a Estrella and
Zubiaga, Irune and
Montejo-R{\'a}ez, Arturo and
Soroa, Aitor and
Mart{\'i}n-Valdivia, Mar{\'i}a Teresa and
Guerini, Marco and
Agerri, Rodrigo",
booktitle = "Proceedings of the First Workshop on Multilingual Counterspeech Generation",
month = jan,
year = "2025",
address = "Abu Dhabi, UAE",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.mcg-1.9/",
pages = "77--91",
abstract = "Hate speech (HS) in online spaces poses severe risks, including real-world violence and psychological harm to victims, necessitating effective countermeasures. Counterspeech (CS), which responds to hateful messages with opposing yet non-hostile narratives, offer a promising solution by mitigating HS while upholding free expression. However, the growing volume of HS demands automation, making Natural Language Processing a viable solution for the automatic generation of CS. Recent works have explored knowledge-driven approaches, leveraging external sources to improve the relevance and informativeness of responses. These methods typically involve multi-step pipelines combining retrieval and passage re-ranking modules. While effective, most studies have focused on English, with limited exploration of multilingual contexts. This paper addresses these gaps by proposing a multilingual, knowledge-driven approach to CS generation. We integrate state-of-the-art re-ranking mechanisms into the CS generation pipeline and evaluate them using the MT-CONAN-KN dataset, which includes hate speech, relevant knowledge sentences, and counterspeech in four languages: English, Italian, Spanish, and Basque. Our approach compares reranker-based systems employing multilingual cross-encoders and LLMs to a simpler end-to-end system where the language model directly handles both knowledge selection and CS generation. Results demonstrate that reranker-based systems outperformed end-to-end systems in syntactic and semantic similarity metrics, with LLM-based re-rankers delivering the strongest performance overall. This work is the result of our participation in the Shared Task on Multilingual Counterspeech Generation held at COLING 2025."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="russo-2025-trenteam">
<titleInfo>
<title>TrenTeam at Multilingual Counterspeech Generation: Multilingual Passage Re-Ranking Approaches for Knowledge-Driven Counterspeech Generation Against Hate</title>
</titleInfo>
<name type="personal">
<namePart type="given">Daniel</namePart>
<namePart type="family">Russo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-01</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the First Workshop on Multilingual Counterspeech Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Helena</namePart>
<namePart type="family">Bonaldi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">María</namePart>
<namePart type="given">Estrella</namePart>
<namePart type="family">Vallecillo-Rodríguez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Irune</namePart>
<namePart type="family">Zubiaga</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Arturo</namePart>
<namePart type="family">Montejo-Ráez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aitor</namePart>
<namePart type="family">Soroa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">María</namePart>
<namePart type="given">Teresa</namePart>
<namePart type="family">Martín-Valdivia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marco</namePart>
<namePart type="family">Guerini</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rodrigo</namePart>
<namePart type="family">Agerri</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Abu Dhabi, UAE</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Hate speech (HS) in online spaces poses severe risks, including real-world violence and psychological harm to victims, necessitating effective countermeasures. Counterspeech (CS), which responds to hateful messages with opposing yet non-hostile narratives, offer a promising solution by mitigating HS while upholding free expression. However, the growing volume of HS demands automation, making Natural Language Processing a viable solution for the automatic generation of CS. Recent works have explored knowledge-driven approaches, leveraging external sources to improve the relevance and informativeness of responses. These methods typically involve multi-step pipelines combining retrieval and passage re-ranking modules. While effective, most studies have focused on English, with limited exploration of multilingual contexts. This paper addresses these gaps by proposing a multilingual, knowledge-driven approach to CS generation. We integrate state-of-the-art re-ranking mechanisms into the CS generation pipeline and evaluate them using the MT-CONAN-KN dataset, which includes hate speech, relevant knowledge sentences, and counterspeech in four languages: English, Italian, Spanish, and Basque. Our approach compares reranker-based systems employing multilingual cross-encoders and LLMs to a simpler end-to-end system where the language model directly handles both knowledge selection and CS generation. Results demonstrate that reranker-based systems outperformed end-to-end systems in syntactic and semantic similarity metrics, with LLM-based re-rankers delivering the strongest performance overall. This work is the result of our participation in the Shared Task on Multilingual Counterspeech Generation held at COLING 2025.</abstract>
<identifier type="citekey">russo-2025-trenteam</identifier>
<location>
<url>https://aclanthology.org/2025.mcg-1.9/</url>
</location>
<part>
<date>2025-01</date>
<extent unit="page">
<start>77</start>
<end>91</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T TrenTeam at Multilingual Counterspeech Generation: Multilingual Passage Re-Ranking Approaches for Knowledge-Driven Counterspeech Generation Against Hate
%A Russo, Daniel
%Y Bonaldi, Helena
%Y Vallecillo-Rodríguez, María Estrella
%Y Zubiaga, Irune
%Y Montejo-Ráez, Arturo
%Y Soroa, Aitor
%Y Martín-Valdivia, María Teresa
%Y Guerini, Marco
%Y Agerri, Rodrigo
%S Proceedings of the First Workshop on Multilingual Counterspeech Generation
%D 2025
%8 January
%I Association for Computational Linguistics
%C Abu Dhabi, UAE
%F russo-2025-trenteam
%X Hate speech (HS) in online spaces poses severe risks, including real-world violence and psychological harm to victims, necessitating effective countermeasures. Counterspeech (CS), which responds to hateful messages with opposing yet non-hostile narratives, offer a promising solution by mitigating HS while upholding free expression. However, the growing volume of HS demands automation, making Natural Language Processing a viable solution for the automatic generation of CS. Recent works have explored knowledge-driven approaches, leveraging external sources to improve the relevance and informativeness of responses. These methods typically involve multi-step pipelines combining retrieval and passage re-ranking modules. While effective, most studies have focused on English, with limited exploration of multilingual contexts. This paper addresses these gaps by proposing a multilingual, knowledge-driven approach to CS generation. We integrate state-of-the-art re-ranking mechanisms into the CS generation pipeline and evaluate them using the MT-CONAN-KN dataset, which includes hate speech, relevant knowledge sentences, and counterspeech in four languages: English, Italian, Spanish, and Basque. Our approach compares reranker-based systems employing multilingual cross-encoders and LLMs to a simpler end-to-end system where the language model directly handles both knowledge selection and CS generation. Results demonstrate that reranker-based systems outperformed end-to-end systems in syntactic and semantic similarity metrics, with LLM-based re-rankers delivering the strongest performance overall. This work is the result of our participation in the Shared Task on Multilingual Counterspeech Generation held at COLING 2025.
%U https://aclanthology.org/2025.mcg-1.9/
%P 77-91
Markdown (Informal)
[TrenTeam at Multilingual Counterspeech Generation: Multilingual Passage Re-Ranking Approaches for Knowledge-Driven Counterspeech Generation Against Hate](https://aclanthology.org/2025.mcg-1.9/) (Russo, MCG 2025)
ACL