Improving Hate Speech Detection with Self-Attention Mechanism and Multi-Task Learning - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2023

Improving Hate Speech Detection with Self-Attention Mechanism and Multi-Task Learning

Abstract

Hate speech detection is a challenging task of natural language processing. Recently, some works have focused on the use of multiword expressions for hate speech detection. In this paper, we propose to use an auxiliary task to improve hate speech detection: multiword expression identification. Our proposed system, based on multi-task with self-attention, outperforms an MWE-based features state-of-the-art system on four hate speech corpora.
Fichier principal
Vignette du fichier
LTC23-9668-Zampieri-multicolor.pdf (393.76 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-04017250 , version 1 (07-03-2023)

Licence

Identifiers

  • HAL Id : hal-04017250 , version 1

Cite

Nicolas Zampieri, Irina Illina, Dominique Fohr. Improving Hate Speech Detection with Self-Attention Mechanism and Multi-Task Learning. LTC'23 - 10th Language & Technology Conference: Human Language Technologies as a Challenge for Computer Science and Linguistics, Apr 2023, Poznan, Poland. ⟨hal-04017250⟩
85 View
205 Download

Share

Gmail Mastodon Facebook X LinkedIn More