Enhanced attentive convolutional neural networks for sentence pair modeling
作者：
Highlights：
• Three enhanced attention mechanisms to make full use of convolution.
• Incorporating local and interactive context in convolution operation.
• Extracting multi-grained similarity features between sentence pairs.
• Pre-trained representations can be combined with our proposed models easily.
• More economical considering the compromise between efficiency and performance.
摘要
•Three enhanced attention mechanisms to make full use of convolution.•Incorporating local and interactive context in convolution operation.•Extracting multi-grained similarity features between sentence pairs.•Pre-trained representations can be combined with our proposed models easily.•More economical considering the compromise between efficiency and performance.
论文关键词：EACNN,Attentive convolution,Multi-grained similarity,Sentence pair modeling
论文评审过程：Received 17 October 2019, Revised 28 February 2020, Accepted 12 March 2020, Available online 13 March 2020, Version of Record 19 March 2020.
论文官网地址：https://doi.org/10.1016/j.eswa.2020.113384
