Chinese Named Entity Recognition On Msra

Métriques

Résultats

Résultats de performance de divers modèles sur ce benchmark

Nom du modèle	F1	Paper Title	Repository
ERNIE 2.0 Base	93.8	ERNIE 2.0: A Continual Pre-training Framework for Language Understanding
NFLAT	94.55	NFLAT: Non-Flat-Lattice Transformer for Chinese Named Entity Recognition
BERT-MRC	95.75	A Unified MRC Framework for Named Entity Recognition
FLAT+BERT	96.09	FLAT: Chinese NER Using Flat-Lattice Transformer
DiffusionNER	94.91	DiffusionNER: Boundary Diffusion for Named Entity Recognition
LSTM + Lexicon augment	93.5	Simplify the Usage of Lexicon in Chinese NER
FLAT	94.12	FLAT: Chinese NER Using Flat-Lattice Transformer
Glyce + BERT	95.54	Glyce: Glyph-vectors for Chinese Character Representations
FGN	95.64	FGN: Fusion Glyph Network for Chinese Named Entity Recognition
ZEN (Init with Chinese BERT)	95.25	ZEN: Pre-training Chinese Text Encoder Enhanced by N-gram Representations
ERNIE 2.0 Large	95	ERNIE 2.0: A Continual Pre-training Framework for Language Understanding
ZEN (Random Init)	93.24	ZEN: Pre-training Chinese Text Encoder Enhanced by N-gram Representations
Baseline + BS	96.26	Boundary Smoothing for Named Entity Recognition
BERT-CRF (Replicated in AdaSeq)	96.69	Improving Named Entity Recognition by External Context Retrieving and Cooperative Learning
TENER	92.74	TENER: Adapting Transformer Encoder for Named Entity Recognition	-
Lattice	93.18	Chinese NER Using Lattice LSTM
W2NER	96.10	Unified Named Entity Recognition as Word-Word Relation Classification
PIQN	93.48	Parallel Instance Query Network for Named Entity Recognition
BERT-MRC+DSC	96.72	Dice Loss for Data-imbalanced NLP Tasks
CAN-NER Model	92.97	CAN-NER: Convolutional Attention Network for Chinese Named Entity Recognition

0 of 21 row(s) selected.