EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Dialect Bias in Arabic Toxicity Detection Under Controlled Distributional Shifts

Maha Alamri

Electronics · 13.08.2026 · 15 (16) · 3612

DOI: 10.3390/electronics15163612 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 13.08.2026. Verfasst von Maha Alamri. Die Arbeit nennt 66 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Toxicity detection models can behave inconsistently across Arabic dialects, yet the level of training-data imbalance at which dialect-conditioned bias becomes statistically detectable remains unquantified. This paper presents a controlled framework for measuring such bias under class-conditional distributional shift. Four transformer encoders (AraBERT, MARBERT, CAMeLBERT-DA, and XLM-RoBERTa) are fine-tuned on a balanced toxicity corpus comprising texts in the Saudi, Egyptian, and Tunisian dialects across five random seeds, after which the representation of Saudi-dialect texts in the toxic class is experimentally increased at three injection levels, λ∈{0.10,0.20,0.30}, under two designs: one that jointly alters toxic-class composition and class balance, and one that isolates toxic-class dialectal composition. Bias emergence is quantified through a statistical bias-sensitivity threshold BSTB, the lowest injection level at which the Saudi-centered false-positive-rate gap is positive and statistically significant under a permutation test. False-negative-rate disparities are additionally evaluated, while ERASER-based comprehensiveness and sufficiency measures and SHAP scenario analysis serve as exploratory attribution diagnostics. Under the first design, BSTB is reached at λ=0.20 or λ=0.30 in most model–seed runs. Under the second design, no model–seed run reaches BSTB within the tested range. The matched-control analysis detects no significant marker-specific comprehensiveness effect.

Zitieren

APA

Alamri, M. (2026). Dialect Bias in Arabic Toxicity Detection Under Controlled Distributional Shifts. Electronics, 15(16), 3612. https://doi.org/10.3390/electronics15163612

Vancouver

Alamri M. Dialect Bias in Arabic Toxicity Detection Under Controlled Distributional Shifts. Electronics. 2026;15(16):3612. doi:10.3390/electronics15163612

BibTeX
@article{Alamri2026,
  author = {Alamri, Maha},
  title = {Dialect Bias in Arabic Toxicity Detection Under Controlled Distributional Shifts},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {16},
  pages = {3612},
  doi = {10.3390/electronics15163612}
}
Kostenlos starten →