PersianMedQA offers a 20,785-question Persian-English medical QA benchmark and a 41-model evaluation, showing that general closed-weight models beat Persian-specialized and medical-specialized models, and that some questions lose meaning in translation.
Expert Evaluation: The model follows a Western protocol; how- ever, local clinical practice requires urgent biopsy due to high mor- tality risk
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
PersianMedQA: Evaluating Large Language Models on a Persian-English Bilingual Medical Question Answering Benchmark
PersianMedQA offers a 20,785-question Persian-English medical QA benchmark and a 41-model evaluation, showing that general closed-weight models beat Persian-specialized and medical-specialized models, and that some questions lose meaning in translation.