AutoArabic: A Three-Stage Framework for Localizing Video-Text Retrieval Benchmarks
This work addresses a gap in Arabic localization for video-text retrieval, providing a scalable method for creating benchmarks, though it is incremental as it adapts existing techniques to a new language domain.
The paper tackled the lack of Arabic video-text retrieval benchmarks by introducing AutoArabic, a three-stage framework that uses LLMs to translate English benchmarks into Arabic, reducing manual revision by nearly fourfold and producing DiDeMo-AR with 40,144 descriptions, while a CLIP-style baseline showed a moderate 3 percentage point performance gap between Arabic and English variants.
Video-to-text and text-to-video retrieval are dominated by English benchmarks (e.g. DiDeMo, MSR-VTT) and recent multilingual corpora (e.g. RUDDER), yet Arabic remains underserved, lacking localized evaluation metrics. We introduce a three-stage framework, AutoArabic, utilizing state-of-the-art large language models (LLMs) to translate non-Arabic benchmarks into Modern Standard Arabic, reducing the manual revision required by nearly fourfold. The framework incorporates an error detection module that automatically flags potential translation errors with 97% accuracy. Applying the framework to DiDeMo, a video retrieval benchmark produces DiDeMo-AR, an Arabic variant with 40,144 fluent Arabic descriptions. An analysis of the translation errors is provided and organized into an insightful taxonomy to guide future Arabic localization efforts. We train a CLIP-style baseline with identical hyperparameters on the Arabic and English variants of the benchmark, finding a moderate performance gap (about 3 percentage points at Recall@1), indicating that Arabic localization preserves benchmark difficulty. We evaluate three post-editing budgets (zero/ flagged-only/ full) and find that performance improves monotonically with more post-editing, while the raw LLM output (zero-budget) remains usable. To ensure reproducibility to other languages, we made the code available at https://github.com/Tahaalshatiri/AutoArabic.