情报智能体——面向“十五五”的科技情报工作新范式

doi:10.13998/j.cnki.issn1002-1248.24-0666

摘要/Abstract

摘要：

目的/意义 科技情报工作的每次变革与发展均围绕先进信息技术的应用展开。目前，大模型、智能体等先进AI技术正在驱动新一轮科研范式变革，情报机构应该思考科技情报工作范式如何变革，为“十五五”发展规划设计奠定方向。 方法/过程 本文使用客观归纳和理论推理方法，整体从AI技术赋能科学研究的3种驱动方式出发，结合情报内涵，推理归纳AI赋能科技情报工作的主要途径体现在两个方面：情报基础建设（数据生产、信息组织、知识表示）和情报生成（情报计算）。融合大模型技术的智能体展现出了极强、甚至比肩科学家的数据理解能力，据此可推断其已具备赋能情报生成的能力。 结果/结论 构建和应用情报智能体是情报机构面向“十五五”的必然选择。在情报智能体的驱动下，科技情报工作将迈向更高水平的自动化与智能化，将情报专家从繁琐的基础数据处理任务中解放出来，专注于高价值情报的生成与决策支持活动。情报智能体驱动的科技情报工作生态下，智能体集群是核心，内外协同运行：对内，通过集成调度工具、基础数据和基础设施资源，实现数据生产、信息组织、知识表示和情报计算四大核心功能的高度自动化；对外，通过智能体与情报专家及具体情报场景的交互，形成“人与智能体协同”的工作新范式。未来，在规划设计情报智能体落地实施时，应同时关注当前研发阶段的技术适配性以及未来应用阶段的潜在安全风险防控，全力保证情报智能体的高效安全应用。

关键词: 十五五, 情报智能体, 科技情报工作, 新范式, 人与智能体协同

Abstract:

Purpose/Significance Every transformation and development in scientific and technological (S&T) documentation and information services has revolved around the application of advanced information technologies. Currently, cutting-edge AI technologies such as large-scale models and agents are driving a new wave of paradigm shifts in scientific research. Information institutions should consider how the paradigm of S&T documentation and information services should evolve to lay a strategic foundation for the development of the "15th Five-Year Plan" development. Method/Process This study uses objective induction and theoretical reasoning methods. It starts with the three driving modes of AI empowering scientific research and combines them with the essence of information work. The study concludes and summarizes that AI empowers S&T documentation and information services in two main areas: information infrastructure (data production, information organization, and knowledge representation) and information generation (intelligence computation). Agents integrated with large-scale modelling technologies demonstrate exceptional, even scientist-level, data understanding capabilities, suggesting that they are already capable of enabling information generation. [Results/ Conclusions Building and deploying DIS agents is an inevitable choice for information institutions as they prepare for the "15th Five-Year Plan". Driven by DIS agents, S&T documentation and information services will achieve higher levels of automation and intelligence, freeing information professionals from tedious basic data processing tasks and allowing them to focus on generating high-value information and supporting decision making. In the ecosystem of S&T documentation and information services driven by DIS agents, clusters of agents form the core and work together both internally and externally: Internally, DIS agents achieve a high level of automation in four core functions: data production, information organization, knowledge representation, and intelligence computation through the integration of planning tools, basic data and infrastructure resources. Externally, through interactions between agents, information experts, and specific intelligence scenarios, a new working paradigm emerges: "human and multi-agent collaboration". In the future, when planning and designing the implementation of DIS agents, it is essential to focus on both the technical adaptability at the current R&D stage and the potential security risks in future application stages. This ensures the efficient and secure use of DIS agents in S&T documentation and information services.

Key words: 15th Five-Year Plan, documentation and information service agent (DIS agent), human-DIS agent collaboration, S&T documentation and information services, new paradigm

中图分类号: G350

刘细文, 付芸, 韦华楠. 情报智能体——面向“十五五”的科技情报工作新范式[J/OL]. 农业图书情报学报. https://doi.org/10.13998/j.cnki.issn1002-1248.24-0666.

Xiwen LIU, Yun FU, Huanan WEI. DIS Agent: New Paradigm of S&T Documentation and Information Service for the Fifteenth Five-Year Plan[J/OL]. Journal of Library and Information Science in Agriculture. https://doi.org/10.13998/j.cnki.issn1002-1248.24-0666.

图/表 7

表1

图1

图2

图3

图4

图5

图6

参考文献 62

1	MILOJEVIĆ S, SUGIMOTO C R, YAN E J, et al. The cognitive structure of library and information science: Analysis of article title words[J]. Journal of the American society for information science and technology, 2011, 62(10): 1933-1953.
2	OPENAI. Hello GPT-4o[EB/OL]. [2024-05-13].
3	TOUVRON H, LAVRIL T, IZACARD G, et al. LLaMA: Open and efficient foundation language models[J/OL]. arXiv preprint arXiv:2302.13971, 2023.
4	GUO T C, CHEN X Y, WANG Y Q, et al. Large language model based multi-agents: A survey of progress and challenges[J/OL]. arXiv preprint arXiv:2402.01680, 2024.
5	MESSERI L, CROCKETT M J. Artificial intelligence and illusions of understanding in scientific research[J]. Nature, 2024, 627(8002): 49-58.
6	YANG J, LI A, FARAJTABAR M, et al. Learning to incentivize other learning agents[J/OL]. arXiv preprint ArXiv: 2006.06051, 2020.
7	SUN W, YAN L, MA X, et al. Is ChatGPT Good at Search? Investigating Large language models as re-ranking agent[J/OL]. arXiv preprint ArXiv: 2304.09542, 2023.
8	OUYANG S Q, LI L. AutoPlan: Automatic planning of interactive decision-making tasks with large language models[C]//Findings of the Association for Computational Linguistics: EMNLP 2023. Stroudsburg, PA, USA: Association for Computational Linguistics, 2023.
9	XUE S Q, ZHOU F, XU Y, et al. WeaverBird: Empowering financial decision-making with large language model, knowledge base, and search engine[J/OL]. arXiv preprint ArXiv: 2308.05361, 2023.
10	GUAN Y, WANG D, CHU Z, et al. Intelligent virtual assistants with LLM-based process automation[J/OL]. arXiv preprint ArXiv: 2312.06677, 2023.
11	YAO S, YU D, ZHAO J, et al. Tree of thoughts: Deliberate problem solving with large language models[J/OL]. arXiv preprint ArXiv: 2305.10601, 2023.
12	SHINN N, CASSANO F, LABASH B, et al. Reflexion: Language agents with verbal reinforcement learning[C]// Proceedings of the Neural Information Processing Systems, New Orleans, USA: the NIPS Foundation, 2023.
13	LI M H, ZHAO Y X, YU B W, et al. API-bank: A comprehensive benchmark for tool-augmented LLMs[C]//Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing. Stroudsburg, PA, USA: Association for Computational Linguistics, 2023.
14	CHU Z, WANG Y, ZHU F, et al. Professional agents - Evolving large language models into autonomous experts with human-level competencies[J/OL]. arXiv preprint ArXiv: 2402.03628, 2024.
15	李国杰. 智能化科研(AI4R): 第五科研范式[J]. 中国科学院院刊, 2024, 39(1): 1-9.
	LI G J. AI4R: The fifth scientific research paradigm[J]. Bulletin of Chinese academy of sciences, 2024, 39(1): 1-9.
16	TSHITOYAN V, DAGDELEN J, WESTON L, et al. Unsupervised word embeddings capture latent knowledge from materials science literature[J]. Nature, 2019, 571(7763): 95-98.
17	刘细文. 贯彻落实二十大精神, 开创文献情报工作的高质量发展道路[J]. 图书情报工作, 2023, 67(1): 4-8.
	LIU X W. Implementing the spirit of the 20th national congress of CPC and planning the road of high-quality development of library and information service[J]. Library and information service, 2023, 67(1): 4-8.
18	刘细文, 付芸. 数智赋能背景下情报学研究进展[J]. 情报学进展, 2024, 15: 131-71.
	LIU X W, FU Y. The research advancements in information science amidst data & AI empowerment: Data-Driven, model-driven and knowledge discovery[J]. Advances in information science, 2024, 15: 131-71.
19	刘细文, 孙蒙鸽, 王茜, 等. DIKIW逻辑链下GPT大模型对文献情报工作的潜在影响分析[J]. 图书情报工作, 2023, 67(21): 3-12.
	LIU X W, SUN M G, WANG X, et al. Analysis of the potential impact of GPT large model under DIKIW logic chain on documentation and information services[J]. Library and information service, 2023, 67(21): 3-12.
20	孙蒙鸽, 韩涛, 王燕鹏, 等. GPT技术变革对基础科学研究的影响分析[J]. 中国科学院院刊, 2023, 38(8): 1212-1224.
	SUN M G, HAN T, WANG Y P, et al. Impact analysis of GPT technology revolution on fundamental scientific research[J]. Bulletin of Chinese academy of sciences, 2023, 38(8): 1212-1224.
21	刘细文. 情报学范式变革与数据驱动型情报工作发展趋势[J]. 图书情报工作, 2021, 65(1): 4-11.
	LIU X W. Paradigm transformation of library and information science and trends of data-driven information services[J]. Library and information service, 2021, 65(1): 4-11.
22	符福峘. 我国情报学理论体系建设的伟大成就——庆祝我国科技情报事业创建50周年[J]. 情报理论与实践, 2006, 29(4): 385-389.
	FU F H. The great achievement of China in the construction of the theoretical system of information science[J]. Information studies: Theory & application, 2006, 29(4): 385-389.
23	关家麟, 张超. 我国科技信息事业发展的回顾与展望[J]. 情报科学, 2007, 25(1): 1-7.
	GUAN J L, ZHANG C. The review and outlook for scientific and technological information undertaking of China[J]. Information science, 2007, 25(1): 1-7.
24	包昌火, 王秀玲, 李艳. 中国情报研究发展纪实[J]. 情报理论与实践, 2010, 33(1): 1-3.
	BAO C H, WANG X L, LI Y. A record of the development of intelligence analysis in China[J]. Information studies: Theory & application, 2010, 33(1): 1-3.
25	中共中央文献研究室. 建国以来重要文献选编-第九册[M]. 北京: 中央文献出版社, 1994.
26	刘如, 吴晨生, 刘彦君, 等. 中国科技情报工作的传承与发展[J]. 情报学报, 2019, 38(1): 38-45.
	LIU R, WU C S, LIU Y J, et al. History and development of scientific and technical intelligence in China[J]. Journal of the China society for scientific and technical information, 2019, 38(1): 38-45.
27	钱学森. 情报资料、图书、文献和档案工作的现代化及其影响[J]. 档案学通讯, 1979(5): 6-10.
	QIAN X S. Modernization of information materials, books, documentation and archives and their implications[J]. Archives science bulletin, 1979(5): 6-10.
28	霍国庆, 汪冰. 穿越冷战的情报科学史及其启示: 理查兹“情报科学与冷战的结束”评介[J]. 情报科学, 1998, 16(2): 89-95.
	HUO G Q, WANG B. The history of information technology across the cold war and its revelation - The review of Richard's information science and the end of the cold war[J]. Information science, 1998, 16(2): 89-95.
29	刘细文. 中国科学院文献情报中心“十四五”发展思考——基于数据、信息、知识与情报的规划框架设计[J]. 数字图书馆论坛, 2021(5): 12-16.
	LIU X W. On the 14th Five-Year Plan of national science library of Chinese academy of sciences: Frameworks design based on the notion of data-information-knowledge-intelligence[J]. Digital library forum, 2021(5): 12-16.
30	孟广均. 中国科学院文献情报中心近40年发展概况(1978-2018)[J]. 图书馆理论与实践, 2019(8): 18-19, 31.
	MENG G J. An overview of the development of the Chinese academy of sciences in recent 40 years(1978-2018)[J]. Library theory and practice, 2019(8): 18-19, 31.
31	刘细文. 对中国科学院文献情报中心业务改革与创新的初步认识[J]. 图书情报工作, 2002, 46(6): 11.
	LIU X W. A preliminary understanding of the business reform and innovation of the document and information center of Chinese Academy of Sciences[J]. Library and information service, 2002, 46(6): 11.
32	张晓林. 超越图书馆: 寻求变革方向——第77届国际图联大会观感[J]. 图书情报工作, 2011, 55(21): 5-10.
	ZHANG X L. Libraries beyond libraries: Seeking for transformative directions - Notes from IFLA 2011[J]. Library and information service, 2011, 55(21): 5-10.
33	谭晓, 靳晓宏. 数智时代情报流程模型构建研究[J]. 情报理论与实践, 2023, 46(10): 32-39.
	TAN X, JIN X H. Research on the construction of intelligence process model in the digital intelligence era[J]. Information studies: Theory & application, 2023, 46(10): 32-39.
34	周晓英, 陈燕方, 张璐. 中国科技情报事业发展历程与发展规律研究[J]. 科技情报研究, 2019, 1(1): 13-28.
	ZHOU X Y, CHEN Y F, ZHANG L. Research on the development process and law of scientific and technical information career in China[J]. Scientific information research, 2019, 1(1): 13-28.
35	吴晨生, 李辉, 付宏, 等. 情报服务迈向3.0时代[J]. 情报理论与实践, 2015, 38(9): 1-7.
	WU C S, LI H, FU H, et al. Intelligence service towards the era of 3.0[J]. Information studies: Theory & application, 2015, 38(9): 1-7.
36	霍忠文, 阎旭军. “情报”、“Informagence”与“Infotelligence”——科技情报工作科学技术属性再思考[J]. 情报理论与实践, 2002, 25(1): 1-5.
	HUO Z W, YAN X J. "Information", "informagence" & "infotelligence"[J]. Information studies: Theory & application, 2002, 25(1): 1-5.
37	KOJIMA T, GU S S, REID M, et al. Large language models are zero-shot reasoners[J]. Advances in neural information processing systems, 2022, 35: 22199-22213.
38	张晓林. Inside-Out & Outside-In: 图书馆服务社会高质量发展的组合模型[J]. 图书馆杂志, 2024, 43(9): 4-10.
	ZHANG X L. Inside-out & outside-In: A combinatorial model for libraries' support of the high-quality development of the society[J]. Library journal, 2024, 43(9): 4-10.
39	李荣, 吴晨生, 董洁, 等. ChatGPT对开源情报工作的影响及对策[J]. 情报理论与实践, 2023, 46(5): 1-5.
	LI R, WU C S, DONG J, et al. Study on the impact of ChatGPT on open source intelligence work and countermeasures[J]. Information studies: Theory & application, 2023, 46(5): 1-5.
40	GHAFAROLLAHI A, BUEHLER M J. SciAgents: Automating scientific discovery through multi-agent intelligent graph reasoning[J/OL]. arXiv preprint ArXiv: 2409.05556, 2024.
41	WANG H C, FU T F, DU Y Q, et al. Scientific discovery in the age of artificial intelligence[J]. Nature, 2023, 620(7972): 47-60.
42	SUH C, FARE C, WARREN J A, et al. Evolving the materials genome: How machine learning is fueling the next generation of materials discovery[J]. Annual review of materials research, 2020, 50: 1-25.
43	JUMPER J, EVANS R, PRITZEL A, et al. Highly accurate protein structure prediction with AlphaFold[J]. Nature, 2021, 596(7873): 583-589.
44	KITANO H. Nobel Turing Challenge: Creating the engine for scientific discovery[J]. NPJ systems biology and applications, 2021, 7(1): 29.
45	SUFI F. Generative pre-trained transformer (GPT) in research: A systematic review on data augmentation[J]. Information, 2024, 15(2): 99.
46	ZHANG Y, GE F, LI F Y, et al. Prediction of multiple types of RNA modifications via biological language model[J]. IEEE/ACM transactions on computational biology and bioinformatics, 2023, 20(5): 3205-3214.
47	ZHOU G, GAO Z, DING Q, et al. Uni-Mol: A universal 3D molecular representation learning framework[C]//Proceedings of the International Conference on Learning Representations, Kigali, Rwanda: The International Conference on Learning Representations, 2023.
48	NIJKAMP E, RUFFOLO J A, WEINSTEIN E N, et al. ProGen2: Exploring the boundaries of protein language models[J]. Cell systems, 2023, 14(11): 968-978.e3.
49	SANTACROCE M, LU Y, YU H, et al. Efficient RLHF: Reducing the memory usage of PPO[J/OL]. arXiv preprint ArXiv: 2309.00754, 2023.
50	LEE H, PHATALE S, MANSOOR H, et al. RLAIF vs. RLHF: Scaling reinforcement learning from human feedback with AI feedback[C]//Proceedings of the International Conference on Machine Learning, Honolulu, Hawaii, USA: International Machine Learning Society (IMLS), 2023.
51	钱学森. 科技情报工作的科学技术[J]. 兵工情报工作(1987年改名《情报理论与实践》), 1983, 6(6): 3-10.
	QIAN X S. Science and technology of scientific and technological information work[J]. Information studies: Theory & application, 1983, 6(6): 3-10.
52	HIXSON W L. The wizards of langley: Inside the CIA's directorate of science and technology[J]. Journal of American history, 2002, 89(3): 1115-1116.
53	SHERMAN K. Strategic intelligence for American world policy[M]. Princeton, N.J.: Princeton University Press, 1966.
54	LIEW A. DIKIW: Data, information, knowledge, intelligence, wisdom and their interrelationships[J]. Business management dynamics, 2013, 2(10): 49-62.
55	Innovation management - Tools and methods for strategic intelligence management - Guidance: ISO 56006: 2021[S]. International Organization for Standardization [iso], 2021.
56	EVANS J, RZHETSKY A. Machine science[J]. Science, 2010, 329(5990): 399-400.
57	钱力, 刘细文, 张智雄, 等. 科技情报智慧数据: 方法、体系与应用[J]. 情报理论与实践, 2024, 47(1): 12-21.
	QIAN L, LIU X W, ZHANG Z X, Et al. Smart data for scientific and technological information: Method, framework and application[J]. Information studies: Theory & application, 2024, 47(1): 12-21.
58	KING R, ZENIL H. A framework for evaluating the AI-driven automation of science[M]. Paris: OECD Publishing, 2023.
59	HUANG Y. Levels of AI agents: From rules to large language models[J/OL]. arXiv preprint ArXiv: 2405.06643, 2024.
60	LU C, LU C, LANGE R T, et al. The AI scientist: Towards fully automated open-ended scientific discovery[J/OL]. arXiv preprint ArXiv: 2408.06292, 2024
61	Crowdflower. Data science report[R]. 2016. San Francisco, USA: Crowdflower. https://www2.cs.uh.edu/~ceick/UDM/CFDS16.pdf.
62	XI Z H, CHEN W X, GUO X, et al. The rise and potential of large language model based agents: A survey[J/OL]. arXiv preprint ArXiv: 2309.07864, 2023.

[1]	满振良, 王昕玮. 总体国家安全观视域下信息迷雾的防治研究[J]. 农业图书情报学报, 2024, 36(3): 83-91.
[2]	王山, 谭宗颖. 关键核心技术识别赋能新质生产力发展：内在逻辑、现实挑战与实践路径[J]. 农业图书情报学报, 2024, 36(2): 26-35.
[3]	孙坦, 张智雄, 周力虹, 王东波, 张海, 李白杨, 勇素华, 左旺孟, 杨光磊. 人工智能驱动的第五科研范式（AI4S）变革与观察[J]. 农业图书情报学报, 0, (): 4-32.
[4]	蒋智慧, 李轩, 曹高辉. 社交媒体用户数字压力成因及影响路径研究[J]. 农业图书情报学报, 2023, 35(11): 64-76.
[5]	郭鹏睿, 文庭孝. 大语言模型对信息检索系统与用户检索行为影响研究[J]. 农业图书情报学报, 2023, 35(11): 13-22.
[6]	孙坦, 张智雄, 周力虹, 王东波, 张海, 李白杨, 勇素华, 左旺孟, 杨光磊. 人工智能驱动的第五科研范式(AI4S)变革与观察[J]. 农业图书情报学报, 2023, 35(10): 4-32.
[7]	罗雯, 胡志刚. 科研合作中的核心合作者的界定与测算——一种基于H指数的测算方式[J]. 农业图书情报学报, 2023, 35(4): 48-58.
[8]	柴雪飞, 邢飞. 重大突发公共卫生事件下老年人群健康信息需求及其服务研究[J]. 农业图书情报学报, 2023, 35(1): 99-107.
[9]	赵又霖, 曹宏楠. 面向应急管理的政务微博信息交流效率及其影响因素研究[J]. 农业图书情报学报, 2022, 34(9): 72-85.
[10]	万昊, 张福俊, 吕千千. 基于DEA超效率验证的项目同行评议结果有效性研究[J]. 农业图书情报学报, 2022, 34(2): 88-101.
[11]	冯劭华, 昝栋, 苏菊, 张展. 专利视角下全球海洋水产饲料领域发展特征分析与启示[J]. 农业图书情报学报, 2021, 33(12): 71-82.
[12]	韩正彪, 周明峰, 岳杭. 低风险疾病情境下农村居民健康信息规避行为研究[J]. 农业图书情报学报, 2021, 33(11): 4-15.
[13]	初景利, 刘培一, 翁彦琴, 李楠, 闫群, 肖玥. 不同角色对开放获取期刊的认知度与认可度调研与分析[J]. 农业图书情报学报, 2021, 33(9): 4-17.
[14]	艾毓茜, 徐健, 何琳, 戚筠. 面向典籍内容分析的分类体系构建方法研究[J]. 农业图书情报学报, 2021, 33(9): 18-26.
[15]	马晓雯, 何琳, 刘建斌, 李章超, 高丹. 基于Bi-LSTM的古籍事件句触发词分类方法研究[J]. 农业图书情报学报, 2021, 33(9): 27-36.