显式知识注入的任务型对话理解模型  

Task-Oriented Dialogue Understanding with Explicit Knowledge Injection

作　　者：李帅鹏 王平辉[1] 孙望淳 杨阳 杜友田[1] 马小科[2] 杜永杰 LI Shuaipeng;WANG Pinghui;SUN Wangchun;YANG Yang;DU Youtian;MA Xiaoke;DU Yongjie(Ministry of Education Key Laboratory for Intelligent Networks and Network Security,Xi’an Jiaotong University,Xi’an 710049,China;School of Computer Science and Technology,Xidian University,Xi’an 710049,China;Digital Home Network National Engineering Laboratory,Haier Group,Qingdao 266000,China)

机构地区：[1]西安交通大学智能网络与网络安全教育部重点实验室,西安710049 [2]西安电子科技大学计算机科学与技术学院,西安710049 [3]青岛海尔科技有限公司数字家庭网络国家工程研究中心,青岛266000

出　　处：《数据采集与处理》2024年第3期668-677,共10页Journal of Data Acquisition and Processing

基　　金：国家重点研发计划(2021YFB1715600);国家自然科学基金(U22B2019,62272372)。

摘　　要：传统对话理解模型依赖对话历史识别用户意图,由于缺乏丰富的知识信息,对生僻或特有内容的理解能力欠佳。通过隐式编码将知识加入模型的方法将知识注入与模型训练高度绑定,难以适应知识库的更新迭代,也会导致知识噪声,引入无关知识破坏原有语义。为解决上述问题,本文提出一种显式知识注入的多任务学习对话理解模型。将知识以自然语言形式插入到对话文本中,即插即用,满足知识源动态发展的需要;通过对话理解的主任务,关联知识识别的辅助任务,进行多任务学习,减少知识噪声。实验结果表明,与现有方法相比,本文提出的模型在意图识别和语义槽填充任务上的宏F_(1)值分别提升了4.87%和2.09%。Dialogue understanding aims to detect user intent given dialogue history.Due to the lack of domain knowledge,traditional dialogue understanding models fail to understand domain-specific entities.Knowledge-enhanced approaches are proposed to improve model performance with structured knowledge,where the knowledge is implicitly injected with knowledge embeddings.However,knowledge embeddings have to be updated with the update of the knowledge base,which brings extra costs.Besides,existing methods suffer from the knowledge noise and incorporate the context-irrelevant knowledge that changes the semantics of the utterance.To address the above issues,this paper proposes a multi-task learning dialogue understanding model with explicit knowledge injection(K-CAM).K-CAM injects knowledge into the model using natural language knowledge without retraining the model for updated knowledge embeddings.A multi-task learning objective of joint intent detection,slot filling,and relevant knowledge recognition is further proposed to resist the knowledge noise problem.Extensive experimental results show that the proposed model K-CAM achieves a significant improvement of 4.87% and 2.09% in macro F_(1) on the intent detection and slot filling tasks compared to other baselines.

关 键 词：任务型对话理解 知识库 多任务学习 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]