Named Entity Recognition On Conll

Résultats

Résultats de performance de divers modèles sur ce benchmark

Nom du modèle	F1	Paper Title
BiLSTM-CRF+ELMo	93.42	Deep contextualized word representations
LUKE + SubRegWeigh (K-means)	95.27	SubRegWeigh: Effective and Efficient Annotation Weighing with Subword Regularization
Pooled Flair	94.13	CrossWeigh: Training Named Entity Tagger from Imperfect Annotations
Noise-robust Co-regularization + LUKE	95.60	Learning from Noisy Labels for Entity-Centric Information Extraction
LSTM-CRF	91.47	Neural Architectures for Named Entity Recognition
Noise-robust Co-regularization + BERT-large	94.04	Learning from Noisy Labels for Entity-Centric Information Extraction
RoBERTa + SubRegWeigh (K-means)	95.45	SubRegWeigh: Effective and Efficient Annotation Weighing with Subword Regularization
CrossWeigh + Pooled Flair	94.28	CrossWeigh: Training Named Entity Tagger from Imperfect Annotations
CL-KL	94.81	Improving Named Entity Recognition by External Context Retrieving and Cooperative Learning
LUKE（Large）	95.89	LUKE: Deep Contextualized Entity Representations with Entity-aware Self-attention
BiLSTM-CNN-CRF	91.87	End-to-end Sequence Labeling via Bi-directional LSTM-CNNs-CRF

0 of 11 row(s) selected.