System Description Paper
Papers describing the methodology, architecture, and results of systems submitted to the shared tasks.
NLP/CORE Unranked
ArabicNLP2026-IslamicEval 2026
IslamicEval 2026 Shared Task: Fine-grained Hallucination Detection in Arabic Islamic Content
ArabicNLP2026-IslamicEval 2026 is a shared task focused on evaluating natural language processing systems for Islamic texts and Arabic-language content related to Islamic contexts. The event aims to advance research in NLP for religious and cultural domains by providing standardized datasets and benchmarks for tasks such as sentiment analysis, question answering, and text classification in Islamic contexts.
Paper fit
Contribution paths
A strong submission should clearly identify its contribution and evaluate it appropriately.
Research areas in scope
01
Shared Task Tracks
Islamic Text Sentiment AnalysisIslamic Question AnsweringFatwa ClassificationQuranic Verse RetrievalHadith ClassificationArabic Dialect Identification in Islamic ContextsIslamic Named Entity Recognition
Policies worth checking twice
- Submissions must be anonymized for double-blind review.
- Papers must not exceed 8 pages (excluding references).
- Dual submission to other conferences or journals is not allowed during review.
- Authors must include an AI-use statement in their submissions.
- All submissions must be in PDF format and follow the ACL anthology template.
Official sources
Compiled from the official call for papers. The organizers’ pages remain authoritative.
Last verified September 9, 2026