Named Entity Recognition Ner On Bc5Cdr
Métriques
F1
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | F1 |
|---|---|
| gollie-annotation-guidelines-improve-zero | 88.4 |
| optimizing-bi-encoder-for-named-entity | 91.9 |
| Modèle 3 | 90.95 |
| biomedical-named-entity-recognition-at-scale | 89.73 |
| scibert-pretrained-contextualized-embeddings | 88.94 |
| a-robust-and-domain-adaptive-approach-for-low | 87.38 |
| biomedical-named-entity-recognition-at-scale | 89.73 |
| scibert-pretrained-contextualized-embeddings | 88.11 |
| collabonet-collaboration-of-deep-neural | 87.12 |
| improving-named-entity-recognition-by | 90.99 |
| accurate-clinical-and-biomedical-named-entity | 90.89 |
| electramed-a-new-pre-trained-language | 90.03 |
| focusing-on-possible-named-entities-in-active | 86 |
| enhancing-label-consistency-on-document-level | 91.3 |
| linkbert-pretraining-language-models-with | 90.22 |
| universalner-targeted-distillation-from-large | 89.34 |