Skip to main content
. 2024 Dec;158:103007. doi: 10.1016/j.artmed.2024.103007

Table 5.

Baseline results of different language models on the biomedical NER tasks.7

Dataset GPT-3.5 Llama-2 Claude-2
NCBI-disease 33.39 4.58 45.75
BC2GM 31.99 5.95 40.45
BC5CDR-chem 41.25 12.21 58.05
BC5CDR-disease 32.26 5.68 50.13
JNLPBA 31.89 4.30 34.62