AraSTEM, a new native Arabic STEM benchmark of 11,637 multiple-choice questions, yields at most 56% accuracy for the best open-source model tested.
Aya dataset: An open-access collection for multilingual in- struction tuning,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
AraSTEM: A Native Arabic Multiple Choice Question Benchmark for Evaluating LLMs Knowledge In STEM Subjects
AraSTEM, a new native Arabic STEM benchmark of 11,637 multiple-choice questions, yields at most 56% accuracy for the best open-source model tested.