Aratox: a multi-dialect, multi-label arabic dataset and model benchmark for toxicity detection.

Toxic language detection in Arabic remains an underexplored yet critical task, complicated by the language’s dialectal diversity, morphological richness, and the prevalence of overlapping toxicity types. In this paper, AraTox is presented as a large-scale, hand-annotated Arabic dataset designed for...

Descripción completa

Detalles Bibliográficos
Publicado en:Language Resources & Evaluation Vol. 60; no. 2; pp. 1 - 39
Autores principales: Alshargi, Faisal, Abulohoom, Ahmed, Yagi, Sane, Jabr, Furat, Lulu, Leena, Elnagar, Ashraf
Formato: Artículo
Publicado: Springer Nature Jun2026
Acceso en línea:Ver este registro en EBSCOhost
fields @attributes:
  recordID: 1
pdfLink:
plink: https://search.ebscohost.com/login.aspx?direct=true&db=hlh&AN=193074424&site=ehost-live
header:
  @attributes:
    shortDbName: hlh
    uiTerm: 193074424
    longDbName: Humanities International Complete
    uiTag: AN
  controlInfo:
    bkinfo:
    jinfo:
      jid:
        1574020X
        179V
      jtl: Language Resources & Evaluation
      issn: 1574020X
      maglogo: N
    pubinfo:
      dt: Jun2026
      vid: 60
      iid: 2
      pid: 237
      pub: Springer Nature
    artinfo:
      ui:
        193074424
        10.1007/s10579-026-09917-9
      ppf: 1
      ppct: 38
      formats:
      tig:
        atl: Aratox: a multi-dialect, multi-label arabic dataset and model benchmark for toxicity detection.
      aug:
        au:
          Alshargi, Faisal
          Abulohoom, Ahmed
          Yagi, Sane
          Jabr, Furat
          Lulu, Leena
          Elnagar, Ashraf
        affil:
          Amazon Robotics, New York, NY, USA
          https://ror.org/00engpz63 Department of Computer Science, University of Sharjah, 27272, University City, Sharjah, UAE
          https://ror.org/00engpz63 Department of Foreign Languages, University of Sharjah, 27272, University City, Sharjah, UAE
          https://ror.org/05k89ew48 Department of English, University of Jordan, Queen Rania St., Amman, Jordan
          https://ror.org/00b98jc42 Department of Computer Science, University of AL Dhaid, AlDhaid, Sharjah, UAE
      sug:
      keyword:
        Annotated dataset
        Arabic NLP
        Dialectal arabic
        Meta-learning
        Multi-label classification
        Toxicity detection
      ab: Toxic language detection in Arabic remains an underexplored yet critical task, complicated by the language’s dialectal diversity, morphological richness, and the prevalence of overlapping toxicity types. In this paper, AraTox is presented as a large-scale, hand-annotated Arabic dataset designed for multi-label and multi-dialect toxicity classification. The dataset comprises around 36,700 social media comments spanning Gulf, Levantine, Nile Basin, North African, Yemeni, and Modern Standard Arabic (MSA) varieties, each manually labeled by expert annotators across seven categories, including hatred, cussing, racial, appearance-based, and sexual insults. The categories were derived from an iterative refinement of overlapping harassment types. AraTox is benchmarked using a range of deep learning models and classical classifiers, highlighting the superior performance of a stacked meta-learning ensemble over standalone models. The meta-classifier achieves a macro-averaged F1-score of 96% and a subset accuracy of 89.67%, significantly outperforming the Multilingual Arabic BERT, MARBERTv2, and other Bidirectional Encoder Representations from Transformers (BERT)-based baselines; i.e., the meta-classifier achieved +5.31% increase in subset accuracy and +5.23% improvement in precision over the strongest deep learning baseline (MARBERTv2). State-of-the-art large language models (LLMs), including GPT-4 and DeepSeek, are further evaluated in zero- and few-shot settings, demonstrating how they struggle with exact-match accuracy (subset accuracy) and exhibit label-set fragmentation. In addition to quantitative evaluation, a qualitative analysis of representative samples is conducted to identify key challenges in annotating and detecting nuanced forms of toxicity. AraTox sets a new benchmark for Arabic toxicity detection and provides a valuable resource for advancing safer and more inclusive Arabic-language moderation systems. Rather than proposing a novel architecture, this work provides a high-utility benchmark and a multi-label dataset. This includes classical learners, fine-tuned Arabic transformers, a stacked ensemble optimized for multi-label classification, and zero-/few-shot evaluation using LLMs. AraTox is released as an open-source resource to foster collaboration and innovation in the Arabic NLP community.
      pubtype: Academic Journal
      doctype: Article
      src: R
    language: English
    refInfo:
    copyright:
      @attributes:
        flag: Y
      dt:
        @attributes:
          year: 2026
    holdings:
      @attributes:
        islocal: N