图情领域大模型的研究进展和未来展望

• 作者： 丁倪 ¹
• 作者单位：
  1.  华东理工大学
• 发布时间：2025-03-10

摘要: [目的/意义]针对大模型的研究呈爆发式增长趋势，有必要对图情领域现有研究进行梳理，探析其研究热点和未来发展方向，对促进生成式人工智能情境下图情学科的创新研究具有重要意义。[方法/过程] 在梳理大模型的技术发展历程的基础上，运用 CiteSpace 软件及文献计量方法，分析图情领域大模型的研究热点，并根据现有研究成果和局限，对未来研究方向进行展望。[结果/结论] 研究结果显示，图情领域在大模型的技术解读与性能测评、跨学科影响与实践路径、用户行为与AI素养、伦理思考与风险应对等四个方面取得较大进展。多模型比较与测试场景拓宽、实践探索与垂直领域优化、用户持续性使用与实践指导、风险研究的多学科融合是未来需要关注的重要问题。

Abstract: [Purpose/Significance] Given the exponential growth in research on large models, it is imperative to systematically review existing studies in the field of library and information science (LIS) to identify current research hotspots and explore future directions. This endeavor holds significant importance for advancing innovative research in LIS within the context of generative artificial intelligence. [Method/Process] Building on a comprehensive review of the technological evolution of large models, this study utilizes CiteSpace software and bibliometric methods to analyze research hotspots of large models in LIS. Furthermore, based on existing research achievements and limitations, it provides a forward-looking perspective on potential future research directions. [Results/Conclusion] The findings indicate that significant progress has been made in LIS across four key areas: technical interpretation and performance evaluation, interdisciplinary impacts and practical pathways, user behavior and AI literacy, ethical considerations and risk management. Moving forward, critical issues that warrant attention include multi-model comparisons and the expansion of testing scenarios, practical exploration and optimization in vertical domains, sustained user engagement and practical guidance, the interdisciplinary integration of risk research.

• 来自： 丁倪
• 分类： 信息资源管理
• 稿件状态： 已投稿期刊
• 引用： PSSXiv:202503.00720 (或此版本 PSSXiv:202503.00720V1)
  DOI:10.12451/202503.00720
  CSTR:32012.36.PSSXiv.202503.00720
• 推荐引用方式： 丁倪.图情领域大模型的研究进展和未来展望.哲学社会科学预印本平台:https://zsyyb.cn/abs/202503.00720.[PSSXiv:202503.00720V1] (点此复制)