Table 5.

P values of the different models for the Chemical-Disease Relations corpus.

Model	Model, P value
	BERTa	BioBERT	EhrBERT500kb	EhrBERT1Mc
DNormd	.004	<.001	<.001	<.001
BERT		.18	.22	.04
BioBERT			.41	.03
EhrBERT500k				.03

^aBERT: bidirectional encoder representations from transformers.

^bEhrBERT_500k: BERT-based model that was trained using 500,000 electronic health record notes.

^cEhrBERT_1M: BERT-based model that was trained using 1 million electronic health record notes.

^dDNorm: disease name normalization.