科研智能关键技术研究

doi:10.12267/j.issn.2096-5931.2025.08.006

摘要/Abstract

摘要：

“人工智能+科研”正重塑科学研究与产业研发范式,开启科研方法论的新纪元。围绕科研智能技术体系,系统梳理科研数据、科研算力、科研模型、科研智能体及自动化实验室五大核心领域,分析其定义内涵、发展现状及未来趋势,探讨人工智能如何驱动科研模式向自主智能演进。研究表明,科研数据治理、异构算力管理、专用及多模态科研模型、智能体自主科研能力和自动化实验室广域协作的深度融合,将支撑形成“想象—执行”闭环的智能科研新范式,为科技成果的高效产出与产业转化提供强劲动力。

关键词: AI, 科研智能, AI4R&D, 科学智能, AI4S

Abstract:

The integration of Artificial Intelligence (AI) with research and development is reshaping the paradigms of both scientific inquiry and industrial research and development, heralding a new era of research methodologies. This paper systematically examines five core technological areas within the "AI for research and development" framework, namely research data, research computing power, research models, research agents, and automated laboratories. It explores their definitions, connotations, current development status, and future trends, while discussing how AI drives scientific research toward autonomous intelligence. The findings indicate that the convergence of research data governance, heterogeneous computing resource management, specialized and multimodal research models, the autonomous research capabilities of intelligent agents, and the wide-area collaboration of automated laboratories collectively supports a novel intelligent research paradigm characterized by an "imagination-execution" closed loop. This paradigm provides robust momentum for accelerating scientific discoveries and facilitating their industrial translation.

Key words: AI, artificial intelligence for research and development, AI4R&D, artificial intelligence for science, AI4S

中图分类号:

TP18
G350.7

董昊, 周景才. 科研智能关键技术研究[J]. 信息通信技术与政策, 2025, 51(8): 42-49.

DONG Hao, ZHOU Jingcai. Research on key technologies of artificial intelligence for research and development[J]. Information and Communications Technology and Policy, 2025, 51(8): 42-49.

导出引用管理器 EndNote|Ris|BibTeX

链接本文:

http://ictp.caict.ac.cn/CN/10.12267/j.issn.2096-5931.2025.08.006

http://ictp.caict.ac.cn/CN/Y2025/V51/I8/42

图/表 1

参考文献 33

[1]	中国人工智能产业发展联盟. 2024科研智能(AI4R&D):人工智能驱动的研发新范式[R], 2024.
[2]	HEY T. The fourth paradigm: data-intensive scientific discovery[M]//KURBANOĞLU S, AL U, ERDOĞAN P L, et al. E-Science and Information Management: Communications in Computer and Information Science. Berlin: Springer, 2012.
[3]	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.
[4]	中国科学技术信息研究所. AI for Science创新图谱[R], 2025.
[5]	JUMPER J, EVANS R, PRITZEL A, et al. Highly accurate protein structure prediction with AlphaFold[J]. Nature, 2021, 596(7873):583-589.
[6]	张婧睿, 孙蒙鸽, 韩涛. 科研智能化趋势下科研数据研究[J]. 科学观察, 2023, 18(4):49-61. doi: 10.15978/j.cnki.1673-5668.202304007
[7]	NOACK M M, ZWART P H, USHIZIMA D M, et al. Gaussian processes for autonomous data acquisition at large-scale synchrotron and neutron facilities[J]. Nature Reviews Physics, 2021,3:685-697.
[8]	廖方宇, 汪洋, 曹荣强, 等. 第五科研范式下新型科研信息化基础平台架构与关键技术[J]. 中国科学院院刊, 2024, 39(12):2048-2059.
[9]	SZYMANSKI N J, RENDY B, FEI Y X, et al. An autonomous laboratory for the accelerated synthesis of novel materials[J]. Nature, 2023, 624(7990):86-91.
[10]	National Center for Biotechnology Information. Current genbank release notes[EB/OL]. (2025-06-15)[2025-06-20]. https://www.ncbi.nlm.nih.gov/genbank/release/current/.
[11]	Semantic Scholar. Semantic scholar academic graph API[EB/OL]. 2025[2025-06-20]. https://www.semanticscholar.org/product/api.
[12]	OpenAlex. What content does OpenAlex include and how does it compare to other databases?[EB/OL]. 2025[2025-06-20]. https://help.openalex.org/hc/en-us/articles/27190301279127-What-content-does-OpenAlex-include-and-how-does-it-compare-to-other-databases.
[13]	AMiner. AMiner开放文档[EB/OL]. 2025[2025-06-20]. https://www.aminer.cn/open/docs.
[14]	TAYLOR R, KARDAS M, CUCURULL G, et al. Galactica: a large language model for science[J]. arXiv Preprint, arXiv:2211.09085, 2022.
[15]	ZHANG D, HU Z, ZHOUBIAN S, et al. SciGLM: training scientific language models with self-reflective instruction annotation and tuning[J]. arXiv Preprint, arXiv:2401.07950, 2024.
[16]	LUO R, SUN L, XIA Y, et al. BioGPT: generative pre-trained transformer for biomedical text generation and mining[J]. Briefings in Bioinformatics, 2022, 23(6):bbac409.
[17]	SINGHAL K, TU T, GOTTWEIS J, et al. Toward expert-level medical question answering with large language models[J]. Nature Medicine, 2025,31:943-950.
[18]	ZHAO Z, MA D, CHEN L, et al. ChemDFM: a large language foundation model for chemistry[J]. arXiv Preprint, arXiv:2401.14818, 2024.
[19]	ZHANG D, LIU W, TAN Q, et al. CheMLLM: a chemical large language model[J]. arXiv Preprint, arXiv:2402.06852, 2024.
[20]	BI K, XIE L, ZHANG H, et al. Accurate medium-range global weather forecasting with 3D neural networks[J]. Nature, 2023, 619(7970):533-538.
[21]	ZHANG K, ZHOU R, ADHIKARLA E, et al. A generalist vision-language foundation model for diverse biomedical tasks[J]. Nature Medicine, 2024,30:3129-3141.
[22]	ZHANG S, XU Y, USUYAMA N, et al. Biomedclip: a multimodal biomedical foundation model pretrained from fifteen million scientific image-text pairs[J]. arXiv Preprint, arXiv:2303.00915, 2023.
[23]	MORO V, LOH C, DANGOVSKI R, et al. Multimodal foundation models for material property prediction and discovery[J]. arXiv Preprint, arXiv:2312.00111, 2025.
[24]	GUO X, LAO J, DANG B, et al. SkySense: a multi-modal remote sensing foundation model towards universal interpretation for earth observation imagery[C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle: IEEE, 2024:27672-27683.
[25]	LU C, LU C, LANGE R T, et al. The AI scientist: towards fully automated open-ended scientific discovery[J]. arXiv Preprint, arXiv:2408.06292, 2024.
[26]	GOTTWEIS J, WENG W H, DARYIN A, et al. Towards an AI co-scientist[J]. arXiv Preprint, arXiv:2502.18864, 2025.
[27]	BRAN M A, COX S, SCHILTER O, et al. Augmenting large language models with chemistry tools[J]. Nature Machine Intelligence, 2024, 6(5):525-535. doi: 10.1038/s42256-024-00832-8 pmid: 38799228
[28]	QIU H, ZHAO J, JING E, et al. Introducing polySea: an LLM-based pymer smart evolution agent[J]. ChemRxiv, 2025.
[29]	LI N, GAO C, LI M, et al. EconAgent: large language model-empowered agents for simulating macroeconomic activities[J]. arXiv Preprint, arXiv:2310.10436, 2023.
[30]	PIAO J, YAN Y, ZHANG J, et al. AgentSociety: large-scale simulation of LLM-driven generative agents advances understanding of human behaviors and society[J]. arXiv Preprint, arXiv:2502.08691, 2025.
[31]	BURGER B, MAFFETTONE P M, GUSEV V V, et al. A mobile robotic chemist[J]. Nature, 2020, 583(7815):237-241.
[32]	ZHU Q, ZHANG F, HUANG Y, et al. An all-round AI-Chemist with a scientific mind[J]. National Science Review, 2022, 9(10):nwac190.
[33]	STRIETH-KALTHOFF F, HAO H, RATHORE V, et al. Delocalized, asynchronous, closed-loop discovery of organic laser emitters[J]. Science, 2024, 384(6697):eadk9227.