📃 Paper • 🤗 HuggingFace
English
Check out HuggingFace datasets to load our data as follows:
from datasets import load_dataset
ds = load_dataset("KenLuo/EMPEC")
EMPEC consists of 157,803 exam questions across 124 subjects and 20 healthcare professions
{
"subject": "解剖學與生理學",
"subject_en": "Anatomy and Physiology",
"profession": "職能治療師",
"profession_en": "Occupational Therapist",
"question": "肺臟中進行氣體交換的主要結構稱做:
A.肺泡(alveolus)
B.氣管(trachea)
C.支氣管(bronchus)
D.橫膈(diaphragm)",
"question_en": "The main structure in the lungs where gas exchange takes place is called:
A. Alveolus
B. Trachea
C. Bronchus
D. Diaphragm",
"answer": "A",
},
python eval.py MODEL_NAME DATA
python test.py eval_results/MODEL_NAME-DATA.jsonl
python eval_api.py MODEL_NAME DATA
python test.py eval_results/MODEL_NAME-DATA.jsonl
Please use the following citation if you intend to use our dataset for training or evaluation:
@article{Luo2024AreLL,
title={Are Large Language Models True Healthcare Jacks-of-All-Trades? Benchmarking Across Health Professions Beyond Physician Exams},
author={Zheheng Luo and Chenhan Yuan and Qianqian Xie and Sophia Ananiadou},
journal={ArXiv},
year={2024},
volume={abs/2406.11328},
url={https://api.semanticscholar.org/CorpusID:270560512}
}