摘要
在人工智能(AI)技术不断发展的背景下,大语言模型(LLM)如OpenAI的GPT系列等已经引发了广泛关注,在包括图书情报在内的许多领域都有潜在应用价值。文章探讨了国内外LLM在文献分类、文摘生成、读者推荐等图书情报任务中的成功应用案例,在分析LLM应用潜力的基础上,进一步分析了知识整合、结构化数据利用、学术真实性检验、AI科学家等LLM应用新方向。但LLM在图书情报领域的应用仍面临计算资源需求大、泛化能力有限的挑战,因而成功利用LLM有赖于图书情报领域的技术和管理创新,需要谨慎推进。
Against the backdrop of continuous development in artificial intelligence(AI)technology,large language models(LLMs)such as OpenAI’s GPT series have gained widespread attention and demonstrated potential applications in various fields,including library and information science.This article explores successful application cases of LLMs in library and information tasks such as literature classification,abstract generation,and reader recommendations both at home and abroad.Building upon the analysis of LLM application potential,it further discusses new directions for LLM applications including knowledge integration,structured data utilization,academic authenticity verification,and AI scientist roles.However,challenges such as significant computational resource requirements and limited generalization capabilities remain for LLMs in library and information science applications.Successful utilization of LLMs relies on technological and managerial innovations within the field,which needs cautious advancement.
作者
洪贇
叶鹰
佟彤
Hong Yun;Ye Ying;Tong Tong
出处
《图书馆理论与实践》
CSSCI
2024年第2期72-80,共9页
Library Theory and Practice
关键词
大语言模型
LLM
GPT
图书情报
AI
技术应用
Large Language Model
LLM
GPT
Library and Information Science
AI
Technological Application
