BriefGPT.xyz
Sep, 2017
稀疏注意力神经网络联合训练的序列标记和分类
Jointly Trained Sequential Labeling and Classification by Sparse Attention Neural Networks
HTML
PDF
Mingbo Ma, Kai Zhao, Liang Huang, Bing Xiang, Bowen Zhou
TL;DR
提出一种利用LSTM网络同时学习句子级别分类任务和序列标注任务的模型，通过语义相关性对单词进行加权的稀疏注意力机制，该方法在ATIS和TREC数据集上表现优于基准模型。
Abstract
sentence-level classification
and
sequential labeling
are two fundamental tasks in language understanding. While these two tasks are usually modeled separately, in reality, they are often correlated, for example
→