图书情报领域大模型的应用模式和数据治理  

Data Governance and Application Development of Large Language Models in Library and Information Services

作　　者：刘倩倩 刘圣婴 刘炜[1] Liu Qianqian;Liu Shengying;Liu Wei(Shanghai Library;East China Normal University Library)

机构地区：[1]上海图书馆 [2]华东师范大学图书馆

出　　处：《图书馆杂志》2023年第12期22-35,共14页Library Journal

基　　金：国家社会科学基金重大项目“文化遗产智慧数据资源建设与服务研究”(项目编号:21&ZD334)的研究成果之一。

摘　　要：本文探讨了图书情报领域大语言模型的应用开发与数据治理要求。大语言模型是依赖海量文本数据,经过无监督预训练及有监督标注数据微调而成。领域大模型则是通用大模型经过领域数据的微调而得到,具备解决领域问题的能力,满足领域应用需求。本文首先回顾了生成式人工智能的突破历程,介绍了大模型的基本原理和应用现状,分析了大模型所具备的多任务能力背后的数据因素和数据需求。最后从数据治理角度重点讨论了领域大模型的应用潜力和方法流程。本文的主要贡献在于分析了图书情报领域大模型的应用模式和数据治理,为图书馆行业应用生成式人工智能技术提供了理论依据和实践指导。同时,文章也讨论了行业大模型应用和评估时需要关注的问题和局限性。This article primarily discusses the data governance requirements and development patterns of large language models in the field of library and information science.Large language models rely on massive amounts of text data for unsupervised pre-training and supervised fine-tuning.Domain-specific large models,on the other hand,are models that have been fine-tuned on domain-specific data to possess domain knowledge and solve domain-specific problems to meet the needs of domain applications.The article aims to explore how to better apply generative artificial intelligence to libraries and related industries to promote the development of smart libraries and provide the driving force for high-quality services.The article first reviews the breakthrough progress of generative artificial intelligence,and then introduces the basic principles and current applications of large models,as well as analyzes the data factors and data requirements behind the various task capabilities of large models.Finally,the article discusses the application potential and development patterns of domain-specific large models.The main contribution of this article is to analyze the application patterns and data governance of large models in the field of library and information services,providing a theoretical basis and practical guidance for the application of generative artificial intelligence technology in the library industry.At the same time,it also discusses the issues and limitations that need to be considered when applying and evaluating industryspecific large models.

关 键 词：大语言模型 领域大模型 大模型应用开发 数据治理 指令微调 图书情报 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] G353.1[自动化与计算机技术—控制科学与工程]