jackkuo/medRxiv_metadata

medRxiv Paper Metadata Dataset

This dataset includes metadata of all papers from medRxiv between June 25, 2019, and March 17, 2025. Number of rows: 69,296 papers.
To access the full text, please visit the official website by searching the paper title or DOI: https://www.medrxiv.org/ , or retrieve the paper using the official API: https://api.medrxiv.org/ .

Dataset Fields

• doi: Digital Object Identifier.
• title: Paper title.
• authors: List of authors.
• author_corresponding: Corresponding author.
• author_corresponding_institution: Corresponding author's institution.
• date: Publication date.
• version: Version information.
• type: Paper type.
• license: License information.
• category: Subject category.
• jatsxml: Content in JATS XML format.
• abstract: Abstract.
• published: Publication status.
• server: Server information.
• index_level_0: Index.

Usage

You can load the dataset using the following code:

import pandas as pd

# Log in using huggingface-cli to access this dataset
df = pd.read_parquet("hf://datasets/jackkuo/medRxiv_meta/combined_medrxiv_data.parquet")

Update Log

• 2025-03-16: Dataset updated; future updates will occur monthly.