Research on Abstractive Text Summarization Based on Encoder Enhancement
HUA Gengxing
ZHU Xinxin
TAO Linjuan
LI Bo
Abstract：The current mainstream abstractive text summarization models are based on the encoder-decoder architecture,where the encoder often uses information from a single source.Based on the recurrent neural network,this paper enhances a single recurrent neural network encoder by incorporating the local semantic information extracted by the convolutional neural network,as well as the topic information extracted by the neural topic model,aiming to improve the performance of text summarization.The ex-perimental results show that the text summarization model using the boosted encoder significantly outperforms the baseline models.
Keywords：abstractive summarizationlocal semantic informationneural topic model
Publication Date：2025-08-20
Online Publishing Date：2025-12-12(First online date of this platform, not the publication date of the document)
Pages：6( 2127-2132 )