.png){kind=icon}
.png){kind=icon}
Return
A two-stage contrastive learning method for nested named entity recognition
.png){kind=icon}
PRE
AI
DOI：10.1016/j.neucom.2026.133382.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A novel two-stage contrastive learning framework for nested named entity recognition. • Boundary-sensitive contrastive learning can effectively improve the accuracy of entity boundary detection. • Four specialized loss functions can effectively enhance the discriminative learning of nested entities. • The model achieves new state-of-the-art performance on benchmark datasets.
Keywords:
nested named entity recognition
contrastive learning
boundary-sensitive learning
loss functions
state-of-the-art performance