Datasets:

Modalities:
Tabular
Text
Formats:
parquet
Languages:
Korean
ArXiv:
Libraries:
Datasets
pandas
License:
sean0042 commited on
Commit
aaff3b5
1 Parent(s): de71ca0

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +16 -3
README.md CHANGED
@@ -217,6 +217,17 @@ We make our data publicly available on HuggingFace and provide a evaluation scri
217
 
218
  Paper : https://arxiv.org/abs/2403.01469
219
 
 
 
 
 
 
 
 
 
 
 
 
220
  ## Dataset Details
221
 
222
  ### Languages
@@ -230,15 +241,17 @@ from datasets import load_dataset
230
  doctor = load_dataset(path = "sean0042/KorMedMCQA",name = "doctor")
231
  nurse = load_dataset(path = "sean0042/KorMedMCQA",name = "nurse")
232
  pharmacist = load_dataset(path = "sean0042/KorMedMCQA",name = "pharm")
 
233
  ```
234
 
235
  ### Statistics
236
 
237
  | Category | # Questions (Train/Dev/Test) |
238
  |------------------------------|------------------------------|
239
- | Doctor | 2,339 (1,890/164/285) |
240
- | Nurse | 1,460 (582/291/587) |
241
- | Pharmacist | 1,546 (632/300/614) |
 
242
 
243
  ### Data Fields
244
 
 
217
 
218
  Paper : https://arxiv.org/abs/2403.01469
219
 
220
+ ## Notice
221
+
222
+ In collaboration with [Hugging Face](https://huggingface.co/datasets/ChuGyouk/KorMedMCQA_edited), we have made the following updates to the KorMedMCQA dataset:
223
+
224
+ 1. **Dentist Exam**: Incorporated exam questions from 2021 to 2024.
225
+ 2. **Updated Test Sets**: Added the 2024 exam questions for the doctor, nurse, and pharmacist test sets.
226
+ 3. **Few-Shot Split**: Introduced a "few-shot" split, containing 5 shots from each validation set.
227
+
228
+ Note that CoT reasoning examples(verified clinicians) for fewshot will be added in [LM-Harness](https://github.com/EleutherAI/lm-evaluation-harness/tree/main/lm_eval/tasks/kormedmcqa) soon. Please stay tuned!
229
+
230
+
231
  ## Dataset Details
232
 
233
  ### Languages
 
241
  doctor = load_dataset(path = "sean0042/KorMedMCQA",name = "doctor")
242
  nurse = load_dataset(path = "sean0042/KorMedMCQA",name = "nurse")
243
  pharmacist = load_dataset(path = "sean0042/KorMedMCQA",name = "pharm")
244
+ dentist = load_dataset(path = "sean0042/KorMedMCQA",name = "dentist")
245
  ```
246
 
247
  ### Statistics
248
 
249
  | Category | # Questions (Train/Dev/Test) |
250
  |------------------------------|------------------------------|
251
+ | Doctor | 2,489 (1,890/164/435) |
252
+ | Nurse | 1,751 (582/291/878) |
253
+ | Pharmacist | 1,817 (632/300/885) |
254
+ | Dentist | 1,412 (297/304/811) |
255
 
256
  ### Data Fields
257