2026
ArabicNLP 2026
ArGuard: Harmful Content Detection in Arabic Memes and LLM Prompts
A two-track shared task covering multimodal hateful meme understanding and harmful Arabic prompt detection for LLM safety evaluation.
Shared Task Archive
ArAIEval
Arabic AI Tasks Evaluation
A consolidated entry point for ArAIEval shared tasks, including Arabic propaganda detection, persuasion techniques, disinformation, multimodal memes, and content safety.
Overview
Evaluation resources for Arabic NLP and multimodal safety
ArAIEval organizes shared tasks and releases datasets, scorer scripts, baselines, and task websites for Arabic language technology evaluation. This page points participants to the current task sites and archival repositories.
2022
WANLP propaganda detection
2023
ArabicNLP ArAIEval
2024
ArabicNLP multimodal propaganda
2026
ArGuard content safety
Shared Tasks
Organized over the years
Each entry links to the task website and repository when available. Older material is kept as an archive so participants can reproduce experiments and reuse evaluation scripts.
2024
ArabicNLP 2024
ArAIEval: Propagandistic Techniques Detection in Unimodal and Multimodal Arabic Content
Shared tasks for text span-level persuasion technique detection and multimodal propagandistic meme classification.
2023
ArabicNLP 2023
ArAIEval: Persuasion Techniques and Disinformation Detection in Arabic Text
Two-task evaluation covering persuasion technique detection in Arabic text and disinformation detection with binary and multiclass subtasks.
2022
WANLP 2022
Propaganda Detection in Arabic
Shared task on identifying Arabic propaganda techniques in tweets, including multilabel technique classification and span-level technique detection.
Contact
ArAIEval coordination
For participant questions, task updates, and archive corrections, contact the organizers or follow the relevant repository.