---
language:
- en
tags:
- biomedical
- bionlp
- entity linking
- embedding
- bert
---
The GEBERT model pre-trained with GAT graph encoder.

The model was published at [CLEF 2023 conference](https://clef2023.clef-initiative.eu/). The source code is available at [github](https://github.com/Andoree/GEBERT).


Pretraining data: biomedical concept graph and concept names from the UMLS (2020AB release).

Base model: [microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext](https://huggingface.co./microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext).



```bibtex
@inproceedings{sakhovskiy2023gebert,
author="Sakhovskiy, Andrey
and Semenova, Natalia
and Kadurin, Artur
and Tutubalina, Elena",
title="Graph-Enriched Biomedical Entity Representation Transformer",
booktitle="Experimental IR Meets Multilinguality, Multimodality, and Interaction",
year="2023",
publisher="Springer Nature Switzerland",
address="Cham",
pages="109--120",
isbn="978-3-031-42448-9"
}
```