A simple and efficient text matching model based on deep interaction

作者：

Highlights：

• We propose a novel model, namely Deep Interaction Text Matching (DITM).

• The proposed model can well capture the interaction information.

• This approach outperforms most of the state-of-the-art methods on multiple tasks.

• The model is simple and effective, with a high generalization ability.

• The study is of great significance to promote the practice of text matching.

摘要

•We propose a novel model, namely Deep Interaction Text Matching (DITM).•The proposed model can well capture the interaction information.•This approach outperforms most of the state-of-the-art methods on multiple tasks.•The model is simple and effective, with a high generalization ability.•The study is of great significance to promote the practice of text matching.

论文关键词：Text matching,Deep learning,Deep interaction,Attention mechanism,Neural network

论文评审过程：Received 1 March 2021, Revised 2 July 2021, Accepted 1 September 2021, Available online 9 September 2021, Version of Record 9 September 2021.

论文官网地址：https://doi.org/10.1016/j.ipm.2021.102738