Datasets:

IIRL-NotreDame
/

ADRD-Bench

Tasks:

Question Answering

Modalities:

Formats:

Languages:

Size:

ArXiv:

Libraries:

License:

Dataset card Data Studio Files Files and versions

Xavier-ZGX commited on 15 days ago

Commit

2a85b8e

·

verified ·

1 Parent(s): 44db867

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -37,7 +37,7 @@ ADRD-Bench consists of **two complementary components**:
 A curated and unified question set synthesized from **seven established medical QA benchmarks**, designed to evaluate LLMs’ **ADRD-related clinical knowledge**.
-- **Total questions**: 1,352
 - **Data access note**: Due to original dataset licensing restrictions, we do **not** redistribute the full question texts. Instead, we provide:
   - the **source dataset name**
   - the **original index** of each question within its source benchmark

 A curated and unified question set synthesized from **seven established medical QA benchmarks**, designed to evaluate LLMs’ **ADRD-related clinical knowledge**.
+- **Total questions**: 1,438
 - **Data access note**: Due to original dataset licensing restrictions, we do **not** redistribute the full question texts. Instead, we provide:
   - the **source dataset name**
   - the **original index** of each question within its source benchmark