Skip to main navigation Skip to search Skip to main content

Arabic Dialect Detection Using Large Language Models: A Comparative Analysis

  • University of Sharjah

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Scopus citations

Abstract

The Arabic language’s rich dialectal variation presents a significant challenge for natural language processing (NLP), especially due to code-switching, phonetic shifts, and region-specific expressions. This study investigates the performance of state-of-the-art large language models (LLMs)—GPT-4, Falcon-7B-Instruct, and LLaMA-3-8B—in classifying Arabic dialects using the MADAR Corpus, a comprehensive dataset covering city-level and regional dialects. The data was preprocessed through normalization, character standardization, and class balancing to enable consistent evaluation across five major dialect groups: Levant, Nile, Gulf, North Africa, and Iraq. In addition to zero-shot and few-shot evaluations of the general-purpose LLMs, we fine-tuned AraBERT on the balanced MADAR dataset and developed a hybrid system combining AraBERT with GPT-4 as a fallback when confidence was low. AraBERT achieved the highest accuracy (68.00%), outperforming GPT-4 (42.80%), Falcon (20%), and LLaMA (23%). The hybrid approach reached 66.13%, reflecting the potential—and complexity—of combining general and domain-specific models. These findings highlight the strengths of task-specific fine-tuning for dialect detection and reveal the limitations of multilingual LLMs when applied to linguistically diverse and underrepresented dialectal data.

Original languageEnglish
Title of host publication2025 International Conference on New Trends in Computing Sciences, ICTCS 2025
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages454-459
Number of pages6
ISBN (Electronic)9798331523411
DOIs
StatePublished - 2025
Event2025 International Conference on New Trends in Computing Sciences, ICTCS 2025 - Amman, Jordan
Duration: 16 Apr 2025 → 18 Apr 2025

Publication series

Name2025 International Conference on New Trends in Computing Sciences, ICTCS 2025

Conference

Conference2025 International Conference on New Trends in Computing Sciences, ICTCS 2025
Country/TerritoryJordan
CityAmman
Period16/04/25 → 18/04/25

Keywords

  • AraBERT
  • Arabic dialect detection
  • Falcon-7B-Instruct
  • GPT-4
  • LLaMA-3-8B
  • MADAR Corpus
  • large language models (LLMs)

Fingerprint

Dive into the research topics of 'Arabic Dialect Detection Using Large Language Models: A Comparative Analysis'. Together they form a unique fingerprint.

Cite this