@inproceedings{di-bonaventura-etal-2025-detection,
title = "From Detection to Explanation: Effective Learning Strategies for {LLM}s in Online Abusive Language Research",
author = "Di Bonaventura, Chiara and
Siciliani, Lucia and
Basile, Pierpaolo and
Merono Penuela, Albert and
McGillivray, Barbara",
editor = "Rambow, Owen and
Wanner, Leo and
Apidianaki, Marianna and
Al-Khalifa, Hend and
Eugenio, Barbara Di and
Schockaert, Steven",
booktitle = "Proceedings of the 31st International Conference on Computational Linguistics",
month = jan,
year = "2025",
address = "Abu Dhabi, UAE",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.coling-main.141/",
pages = "2067--2084",
abstract = "Abusive language detection relies on understanding different levels of intensity, expressiveness and targeted groups, which requires commonsense reasoning, world knowledge and linguistic nuances that evolve over time. Here, we frame the problem as a knowledge-guided learning task, and demonstrate that LLMs' implicit knowledge without an accurate strategy is not suitable for multi-class detection nor explanation generation. We publicly release GLlama Alarm, the knowledge-Guided version of Llama-2 instruction fine-tuned for multi-class abusive language detection and explanation generation. By being fine-tuned on structured explanations and external reliable knowledge sources, our model mitigates bias and generates explanations that are relevant to the text and coherent with human reasoning, with an average 48.76{\%} better alignment with human judgment according to our expert survey."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="di-bonaventura-etal-2025-detection">
<titleInfo>
<title>From Detection to Explanation: Effective Learning Strategies for LLMs in Online Abusive Language Research</title>
</titleInfo>
<name type="personal">
<namePart type="given">Chiara</namePart>
<namePart type="family">Di Bonaventura</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lucia</namePart>
<namePart type="family">Siciliani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pierpaolo</namePart>
<namePart type="family">Basile</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Albert</namePart>
<namePart type="family">Merono Penuela</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Barbara</namePart>
<namePart type="family">McGillivray</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 31st International Conference on Computational Linguistics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Owen</namePart>
<namePart type="family">Rambow</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Leo</namePart>
<namePart type="family">Wanner</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marianna</namePart>
<namePart type="family">Apidianaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hend</namePart>
<namePart type="family">Al-Khalifa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Barbara</namePart>
<namePart type="given">Di</namePart>
<namePart type="family">Eugenio</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Steven</namePart>
<namePart type="family">Schockaert</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>Abusive language detection relies on understanding different levels of intensity, expressiveness and targeted groups, which requires commonsense reasoning, world knowledge and linguistic nuances that evolve over time. Here, we frame the problem as a knowledge-guided learning task, and demonstrate that LLMs’ implicit knowledge without an accurate strategy is not suitable for multi-class detection nor explanation generation. We publicly release GLlama Alarm, the knowledge-Guided version of Llama-2 instruction fine-tuned for multi-class abusive language detection and explanation generation. By being fine-tuned on structured explanations and external reliable knowledge sources, our model mitigates bias and generates explanations that are relevant to the text and coherent with human reasoning, with an average 48.76% better alignment with human judgment according to our expert survey.</abstract>
<identifier type="citekey">di-bonaventura-etal-2025-detection</identifier>
<location>
<url>https://aclanthology.org/2025.coling-main.141/</url>
</location>
<part>
<date>2025-01</date>
<extent unit="page">
<start>2067</start>
<end>2084</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T From Detection to Explanation: Effective Learning Strategies for LLMs in Online Abusive Language Research
%A Di Bonaventura, Chiara
%A Siciliani, Lucia
%A Basile, Pierpaolo
%A Merono Penuela, Albert
%A McGillivray, Barbara
%Y Rambow, Owen
%Y Wanner, Leo
%Y Apidianaki, Marianna
%Y Al-Khalifa, Hend
%Y Eugenio, Barbara Di
%Y Schockaert, Steven
%S Proceedings of the 31st International Conference on Computational Linguistics
%D 2025
%8 January
%I Association for Computational Linguistics
%C Abu Dhabi, UAE
%F di-bonaventura-etal-2025-detection
%X Abusive language detection relies on understanding different levels of intensity, expressiveness and targeted groups, which requires commonsense reasoning, world knowledge and linguistic nuances that evolve over time. Here, we frame the problem as a knowledge-guided learning task, and demonstrate that LLMs’ implicit knowledge without an accurate strategy is not suitable for multi-class detection nor explanation generation. We publicly release GLlama Alarm, the knowledge-Guided version of Llama-2 instruction fine-tuned for multi-class abusive language detection and explanation generation. By being fine-tuned on structured explanations and external reliable knowledge sources, our model mitigates bias and generates explanations that are relevant to the text and coherent with human reasoning, with an average 48.76% better alignment with human judgment according to our expert survey.
%U https://aclanthology.org/2025.coling-main.141/
%P 2067-2084
Markdown (Informal)
[From Detection to Explanation: Effective Learning Strategies for LLMs in Online Abusive Language Research](https://aclanthology.org/2025.coling-main.141/) (Di Bonaventura et al., COLING 2025)
ACL