Source: MedMCQA: A Large-scale Multi-Subject Multi-Choice Dataset for Medical domain Question Answering

The MedMCQA is a large-scale benchmark dataset of Multiple-Choice Question Answering (MCQA) dataset designed to address real-world medical entrance exam questions.

subsets Details
MedMCQA-Test This dataset does not contain labels, so accuracy and fairness tests cannot be run on it. Only robustness tests can be applied.
MedMCQA-Validation This dataset does contain labels, enabling the execution of robustness, accuracy, and fairness tests.

Both the subset contains the following splits:

  • Anaesthesia
  • Anatomy
  • Biochemistry
  • Dental
  • ENT
  • Forensic_Medicine
  • Gynaecology_Obstetrics
  • Medicine
  • Microbiology
  • Ophthalmology
  • Pathology
  • Pediatrics
  • Pharmacology
  • Physiology
  • Psychiatry
  • Radiology
  • Skin
  • Social_Preventive_Medicine
  • Surgery
  • Unknown
Last updated