EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

TARGE: large language model-powered explainable hate speech detection

Muhammad Haseeb Hashir, Memoona, Sung Won Kim iD

PeerJ Computer Science · 30.05.2025 · 11 · e2911

DOI: 10.7717/peerj-cs.2911 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PeerJ Computer Science, online erschienen am 30.05.2025. Verfasst von Muhammad Haseeb Hashir und 2 weiteren Autor:innen. Beteiligte Einrichtung: Information and Communication Engineering, Yeungnam University, Gyeongsan, Gyeongbuk, Republic of South Korea. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 58 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Basic Science Research Program through the National Research Foundation of Korea, Ministry of Education, Korean government.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The proliferation of user-generated content on social networking sites has intensified the challenge of accurately and efficiently detecting inflammatory and discriminatory speech at scale. Traditional manual moderation methods are impractical due to the sheer volume and complexity of online discourse, necessitating automated solutions. However, existing deep learning models for hate speech detection typically function as black-box systems, providing binary classifications without interpretable insights into their decision-making processes. This opacity significantly limits their practical utility, particularly in nuanced content moderation tasks. To address this challenge, our research explores leveraging the advanced reasoning and knowledge integration capabilities of state-of-the-art language models, specifically Mistral-7B, to develop transparent hate speech detection systems. We introduce a novel framework wherein large language models (LLMs) generate explicit rationales by identifying and analyzing critical textual features indicative of hate speech. These rationales are subsequently integrated into specialized classifiers designed to perform explainable content moderation. We rigorously evaluate our methodology on multiple benchmark English-language social media datasets. Results demonstrate that incorporating LLM-generated explanations significantly enhances both the interpretability and accuracy of hate speech detection. This approach not only identifies problematic content effectively but also clearly articulates the analytical rationale behind each decision, fulfilling the critical demand for transparency in automated content moderation.

Schlagwörter

Stichwörter der Autor:innen

Hate speechLarge language modelsRationale extractionSocial media

Autor:innen und Institutionen

  • Information and Communication Engineering, Yeungnam University, Gyeongsan, Gyeongbuk, Republic of South Korea.
  • School of Computer Science and Engineering, Yeungnam University, Gyeongsan, Gyeongbuk, Republic of South Korea.

ORCID-Profile: Sung Won Kim

Förderung

  • Basic Science Research Program through the National Research Foundation of Korea
  • Ministry of Education
  • Korean government

Zitieren

APA

Hashir, M. H., Memoona, & Kim, S. W. (2025). TARGE: large language model-powered explainable hate speech detection. PeerJ Computer Science, 11, e2911. https://doi.org/10.7717/peerj-cs.2911

Vancouver

Hashir MH, Memoona, Kim SW. TARGE: large language model-powered explainable hate speech detection. PeerJ Computer Science. 2025;11:e2911. doi:10.7717/peerj-cs.2911

BibTeX
@article{Hashir2025,
  author = {Hashir, Muhammad Haseeb and Memoona and Kim, Sung Won},
  title = {TARGE: large language model-powered explainable hate speech detection},
  journal = {PeerJ Computer Science},
  year = {2025},
  volume = {11},
  pages = {e2911},
  doi = {10.7717/peerj-cs.2911}
}
Kostenlos starten →