A knowledge graph-based retrieval-augmented generation framework is proposed to achieve military intelligence ques-tion answering.The framework effectively acquires background knowledge through question classification,entity recognition,en-tity linking,and knowledge retrieval.Considering the multi-constraint characteristics of intelligence questions,answer set pro-gramming is used to reduce the amount of knowledge through constraints or to directly obtain answers.Finally,a large language model solves the questions based on the refined knowledge,minimizing attribute recognition and linking issues during question understanding.Experiments on the MilRE dataset demonstrate that the framework provides enhanced knowledge retrieval capa-bilities based on knowledge graphs and offers superior performance in answering military intelligence questions.
关键词
情报问答/回答集编程/大语言模型/检索增强生成/知识图谱
Key words
Intelligence question-answering/Answer set programming/Large language models/Retrieval-augmented generation/Knowledge graph
维普
万方数据