Improving Implicit Hate Speech Detection via a Community-Driven Multi-Agent Framework

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Gajewska, Ewelina, Budzynska, Katarzyna, Chudziak, Jarosław A
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917224557576192
author Gajewska, Ewelina
Budzynska, Katarzyna
Chudziak, Jarosław A
author_facet Gajewska, Ewelina
Budzynska, Katarzyna
Chudziak, Jarosław A
contents This work proposes a contextualised detection framework for implicitly hateful speech, implemented as a multi-agent system comprising a central Moderator Agent and dynamically constructed Community Agents representing specific demographic groups. Our approach explicitly integrates socio-cultural context from publicly available knowledge sources, enabling identity-aware moderation that surpasses state-of-the-art prompting methods (zero-shot prompting, few-shot prompting, chain-of-thought prompting) and alternative approaches on a challenging ToxiGen dataset. We enhance the technical rigour of performance evaluation by incorporating balanced accuracy as a central metric of classification fairness that accounts for the trade-off between true positive and true negative rates. We demonstrate that our community-driven consultative framework significantly improves both classification accuracy and fairness across all target groups.
format Preprint
id arxiv_https___arxiv_org_abs_2601_09342
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Improving Implicit Hate Speech Detection via a Community-Driven Multi-Agent Framework
Gajewska, Ewelina
Budzynska, Katarzyna
Chudziak, Jarosław A
Computation and Language
Artificial Intelligence
This work proposes a contextualised detection framework for implicitly hateful speech, implemented as a multi-agent system comprising a central Moderator Agent and dynamically constructed Community Agents representing specific demographic groups. Our approach explicitly integrates socio-cultural context from publicly available knowledge sources, enabling identity-aware moderation that surpasses state-of-the-art prompting methods (zero-shot prompting, few-shot prompting, chain-of-thought prompting) and alternative approaches on a challenging ToxiGen dataset. We enhance the technical rigour of performance evaluation by incorporating balanced accuracy as a central metric of classification fairness that accounts for the trade-off between true positive and true negative rates. We demonstrate that our community-driven consultative framework significantly improves both classification accuracy and fairness across all target groups.
title Improving Implicit Hate Speech Detection via a Community-Driven Multi-Agent Framework
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2601.09342