基于大模型的标准文献智能问答技术研究
Research on Intelligent Question Answering for Standard Literature based on Large-Scale Models
程云 1吕爽 1陈国祥1
作者信息
摘要
为了优化标准化工作流程,提高标准化工作效率,推动标准数字化发展,介绍了大语言模型(Large Language Model,LLM)在智能问答中的演进与创新,利用大语言模型和检索增强生成(Retrieval-Augmented Generation,RAG)技术,构建了一个标准文献智能问答解决方案,可通过对标准文档的深入理解和智能化处理,实现对复杂标准问题的准确回答,从而增强标准文献的应用价值和实际效益.
Abstract
To optimize standardized workflows,enhance the efficiency of standardization efforts,and promote the digital development of standards,this article discusses the evolution and innovation of large language models in intelligent question-answering.Utilizing large language models(LLM)and retrieval-augmented generation(RAG)technology,we have developed an intelligent question-answering solution for standard literature.By deeply understanding and intelligently processing standard documents,this solution can accurately answer complex standard-related questions,thereby enhancing the application value and practical benefits of standard literature.
关键词
标准数字化/大语言模型/智能问答系统/检索增强生成Key words
standard digitization/Large Language Model(LLM)/Intelligent Question-Answering System/Retrieval-Augmented Generation(RAG)引用本文复制引用
出版年
2024
NETL
NSTL
万方数据