基于关键词异构图的生成式摘要研究  

KHGAS:Keywords Guided Heterogeneous Graph for Abstract:ive Summarization

作　　者：毛兴静 魏勇[2] 杨昱睿 琚生根[1] MAO Xingjing;WEI Yong;YANG Yurui;JU Shenggen(College of Computer Science,Sichuan University,Chengdu 610065,China;No.30 Research Institute of CETC,Chengdu 610041,China)

机构地区：[1]四川大学计算机学院,成都610065 [2]中国电子科技集团公司第三十研究所,成都610041

出　　处：《计算机科学》2024年第7期278-286,共9页Computer Science

基　　金：国家自然科学基金重点项目(62137001)。

摘　　要：生成式摘要是自然语言处理中的重要任务,它帮助人们从海量文本中提取简洁而重要的信息.目前主流的生成式摘要模型是基于深度学习的序列到序列模型,这类模型生成的摘要质量更高.但由于缺乏对原文中关键词和句子之间的依赖关系的关注,现有模型生成的摘要仍然存在语义不明、重要信息含量低等问题.针对这个问题,提出了一种基于关键词异构图的生成式摘要模型.该模型通过从原始文本中提取关键词,将其与句子共同作为输入构建异构图,进而学习关键词和句子之间的依赖关系.文档编码器和图编码器分别用于学习文本知识和异构图中的依赖关系.此外,在解码器中采用分层图注意力机制来提高模型在生成摘要时对显著信息的关注.在CNN/Daily Mail和XSum数据集上进行了充分的实验,实验结果表明,所提模型在ROUGE评价指标上有了显著的提升.进一步的人类评估结果显示,所提模型所生成的摘要比基线模型包含更多的关键信息,并具有更高的可读性.Abstractive summarization is a crucial task in natural language processing that aims to generate concise and informative summaries from a given text.Deep learning-based sequence-to-sequence models have become the mainstream approach for generating Abstractive summaries,achieving remarkable performance gains.However,existing models still suffer from issues such as semantic ambiguity and low information content due to the lack of attention to the dependency relationships between key concepts and sentences in the input text.To address this challenge,the keywords guided heterogeneous graph model for Abstract:ive summarization is proposed.This model leverages extracted keywords and constructs a heterogeneous graph with both keywords and sentences as input to model the dependency relationships between them.A document encoder and a graph encoder are respectively used to capture textual information and dependency relationships in the heterogeneous graph.Moreover,a hierarchical graph attention mechanism is introduced in the decoder to improve the model’s attention to significant information when generating summaries..Extensive experiments on the CNN/Daily Mail and XSum datasets demonstrate that the proposed model outperforms existing methods in terms of the ROUGE evaluation metric.Human evaluations also reveal that the generated summaries by the proposed model contain more key information and are more readable compared to the baseline models.

关 键 词：生成式摘要 关键词 异构图 图注意力 序列到序列模型 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]