Effectiveness of Counter-Speech against Abusive Content: A Multidimensional Annotation and Classification Study

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Damo, Greta, Cabrio, Elena, Villata, Serena
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912644980539392
author Damo, Greta
Cabrio, Elena
Villata, Serena
author_facet Damo, Greta
Cabrio, Elena
Villata, Serena
contents Counter-speech (CS) is a key strategy for mitigating online Hate Speech (HS), yet defining the criteria to assess its effectiveness remains an open challenge. We propose a novel computational framework for CS effectiveness classification, grounded in linguistics, communication and argumentation concepts. Our framework defines six core dimensions - Clarity, Evidence, Emotional Appeal, Rebuttal, Audience Adaptation, and Fairness - which we use to annotate 4,214 CS instances from two benchmark datasets, resulting in a novel linguistic resource released to the community. In addition, we propose two classification strategies, multi-task and dependency-based, achieving strong results (0.94 and 0.96 average F1 respectively on both expert- and user-written CS), outperforming standard baselines, and revealing strong interdependence among dimensions.
format Preprint
id arxiv_https___arxiv_org_abs_2506_11919
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Effectiveness of Counter-Speech against Abusive Content: A Multidimensional Annotation and Classification Study
Damo, Greta
Cabrio, Elena
Villata, Serena
Computation and Language
Counter-speech (CS) is a key strategy for mitigating online Hate Speech (HS), yet defining the criteria to assess its effectiveness remains an open challenge. We propose a novel computational framework for CS effectiveness classification, grounded in linguistics, communication and argumentation concepts. Our framework defines six core dimensions - Clarity, Evidence, Emotional Appeal, Rebuttal, Audience Adaptation, and Fairness - which we use to annotate 4,214 CS instances from two benchmark datasets, resulting in a novel linguistic resource released to the community. In addition, we propose two classification strategies, multi-task and dependency-based, achieving strong results (0.94 and 0.96 average F1 respectively on both expert- and user-written CS), outperforming standard baselines, and revealing strong interdependence among dimensions.
title Effectiveness of Counter-Speech against Abusive Content: A Multidimensional Annotation and Classification Study
topic Computation and Language
url https://arxiv.org/abs/2506.11919