BAREC-ST-2026. The Second Shared Task on Sentence-level Readability Assessment (Strict Track): An NLP competition on fine-grained Arabic readability classification across 19 levels.
-
Updated
Aug 7, 2026 - Python
BAREC-ST-2026. The Second Shared Task on Sentence-level Readability Assessment (Strict Track): An NLP competition on fine-grained Arabic readability classification across 19 levels.
Ai Application For Enhancing Customer Services
BAREC-ST-2026. The Second Shared Task on Sentence-level Readability Assessment (Open Track): An NLP competition on fine-grained Arabic readability classification across 19 levels.
Production Arabic complaint triage — 3 MARBERT models classify, a rule engine decides the action, an LLM only explains. FastAPI · Docker · Railway.
Arabic stance detection system for StanceEval-2026 on Mawqif-v2, combining fine-tuned Arabic encoders, retrieval-augmented few-shot LLMs, LoRA adapters, and a per-class stacking ensemble for seen and unseen target evaluation.
To associate your repository with the marbert topic, visit your repo's landing page and select "manage topics."