Dataset · ncbi
PubMed biomedical citations
ncbi/pubmed
Annual MEDLINE/PubMed citation baseline in English XML with daily updates.
What's inside
Over 10 million up to 100 million records of MEDLINE/PubMed citations in English. NLM releases an annual baseline version in XML every December along with daily files containing new, changed, and deleted citations. Tasks include text generation, fill-mask, and classification.
What you can do with it
- Training biomedical LLMs
- Article classification
- Fill-mask tasks on medical texts
- Tasks
- Text generation, Fill mask, Text classification
- Languages
- English
- Size
- 10M–100M rows
- Downloads this month
- 715
- Likes
- 174
Models trained on it
- Simonlee711/Clinical_ModernBERTEmbeddings 3.4K
- microsoft/MediPhi-InstructText generation 3K
- microsoft/MediPhiText generation 2.6K
- microsoft/MediPhi-PubMedText generation 2.5K
- mradermacher/MediPhi-i1-GGUF 872
- mradermacher/MediPhi-Instruct-i1-GGUF 667
- gabriellarson/MediPhi-Instruct-GGUF 547
- mradermacher/MediPhi-Instruct-GGUF 312
How to load it
from datasets import load_dataset
ds = load_dataset("ncbi/pubmed")