Reconstructed Arabic-LJP Dataset

Published: 5 January 2026| Version 1 | DOI: 10.17632/48cdh9jkc3.1
Contributor:
Nora mahmoud

Description

This dataset contains the fully reconstructed version of the Arabic-LJP corpus, transformed into a retrieval-oriented benchmark with metadata-enhanced case profiles, teacher-generated candidate pairs, human-validated relevance labels, and supervised triplets for dense retrieval training.

Files

Institutions

  • Kafrelsheikh University

Categories

Textual Database

Licence