BriefGPT.xyz
Aug, 2019
分层细化标签注意力网络用于序列标注
Hierarchically-Refined Label Attention Network for Sequence Labeling
HTML
PDF
Leyang Cui, Yue Zhang
TL;DR
提出一种基于层级注意力机制的标签嵌入模型，较传统的BiLSTM-CRF有更好的标签序列表示性能，能显著提高词性标注、实体识别和语法标记任务的标注效果并缩短训练和测试时间。
Abstract
crf
has been used as a powerful model for statistical sequence labeling. For
neural sequence labeling
, however, BiLSTM-
crf
does not always
→