Sudanese Arabic Dialect Identification Benchmark

A curated benchmark dataset for Arabic dialect identification, with special focus on Sudanese Arabic (ARS) — an under-represented variety in NLP research. This dataset enables evaluation of: Dialect identification models Cross-dialectal transfer learning Low-resource language modeling Arabic NLP robustness testing Task Description