| 1,300 | 0 | 66 |
| 下载次数 | 被引频次 | 阅读次数 |
[目的/意义]为解决当前专利检索模式在检索人员不熟悉专业领域或需优先检索基础专利文献时存在的局限性问题，提出一种基于大模型RAG架构的专利检索Agent及检索模式。[方法/过程]该专利检索Agent支持专利文献库的分级搭建，允许检索人员根据需求和习惯构建个性化的多类型专利文献库，同时采用一种可反馈调整的二次切分向量转化方法提高检索准确度，通过配置Agent的基本功能实现专利检索过程中的互联网数据检索，并借助大模型的生成能力对检索结果进行精炼和输出。[结果/结论]通过两个具体案例，验证该专利检索Agent可实现基础文献库的优先准确检索以及内外网检索的无缝衔接，为提升检索效率以及确保检索结果的准确性提供了全新的视角与策略。
Abstract：[Purpose/Significance] To address the limitations of current patent retrieval models when retrievers are unfamiliar with the professional field or need to prioritize the retrieval of foundational literature, a patent retrieval Agent and retrieval mode based on the large language model retrieval-augmented generation(RAG) architecture are proposed. [Method/Process] This patent retrieval Agent supports the hierarchical establishment of patent literature databases, allowing retrievers to construct personalized multitype patent literature databases according to their needs and habits. It also adopts a feedback-adjustable secondary segmentation vector transformation method to improve retrieval accuracy. By configuring the basic capabilities of the Agent, Internet knowledge retrieval is achieved, and the retrieval results are refined and output with the help of the large model's generative capabilities. [Result/Conclusion] Through two specific cases, it is demonstrated that the patent retrieval Agent can achieve accurate and prioritized retrieval of foundational literature databases and seamless integration of LAN and Internet retrieval, providing a new perspective and strategy for improving retrieval efficiency and ensuring the accuracy of retrieval results.
[1]DOUGLAS M R.Large Language Models[J].Communications of the ACM,2023,66:7.DOI:10.1145/3606337.
[2]The United States Patent and Trademark Office.New artificial intelligence functionality in PE2E Search[EB/OL].(2022-11-15)[2024-02-27].https://www.uspto.gov/site s/default/files/documents/ai-simsearch.pdf.
[3]The European Patent Office.New CPC text categoriser is powered by AI[EB/OL].(2023-11-07)[2024-03-02].https://www.epo.org/e n/news-events/new s/new-cpc-text-c ategoriser-powered-ai.
[4]特許庁.特許庁における人工知能(AI)技術の活用に向けたアクション?プラン(令和4~8年度版)について[EB/OL].(2022-06-24)[2024-03-02].https://www.jpo.go.jp/system/laws/sesaku/ai_action_plan/ai_action_plan-fy2022.html.
[5](?)[EB/OL].(2023-12-21)[2024-03-02].https://www.kipo.go.kr/ko/kpoBultnDetail.do?menuCd=S CD0200618&ntatcSeq=19974&aprchId=BUT0000029&sy sC d=SCD02.
[6]申长雨.开拓创新锐意进取奋力开启知识产权事业发展新征程:国家知识产权局局长申长雨在2024年全国知识产权局局长会议上的工作报告(摘编)[N].中国知识产权报,2024-01-05(3).
[7]申长雨.进一步深化知识产权领域改革为中国式现代化提供有力支撑[J/OL].求是,2024(16)[2024-09-14].http://www.qstheory.cn/dukan/qs/2024-08/16/c_1130192040.htm.
[8]费一楠,张文,邬小倩,等.大模型技术在专利审查中的应用探讨[J].中国发明与专利,2024,21(6):4-12.
[9]梁钦锐,李艳.ChatGPT类人工智能应用对专利工作的影响探讨[J].中国发明与专利,2023,20(9):39-47.
[10]张成鲁,刘宝琪,李晓松,等.大模型赋能国防知识产权工作——作用机理与应用模式[J].中国发明与专利,2024,21(7):4-10.
[11]LEE J S,HSIANG J.Patent classification by finetuning BERT language model[J/OL].World Patent Information,2020,61:101965.DOI:10.1016/j.wpi.2020.101965.
[12]LO H C,CHU J M,HSIANG J,et al.Large language model informed patent image retrieval[J/OL].2024[2024-09-10].https://doi.org/10.48550/arXiv.2404.19360.
[13]RAGOT S.A novel approach to measuring the scope of patent claims based on probabilities obtained from(large)language models[J].World Patent Information,2024,79:102321.
[14]丁俐夫,陈颖,肖谭南,等.基于大模型的新型电力系统生成式智能应用模式初探[J].电力系统自动化,2024,48(19):1-13.
[15]钟逸,冷彦,陈思慧,等.基于大模型RAG架构的电池加速研究:现状与展望[J].储能科学与技术,2024,13(9):3214-3225.
[16]MATSUMOTO N,MORAN J,CHOI H,et al.KRAGEN:a knowledge graph-enhanced RAG framework for biomedical problem solving using large language models[J/OL].Bioinformatics,2024,40(6):353.DOI:10.1093/bioinformatics/btae353.
[17]百度Geek说.百度智能云千帆大模型平台引领企业创新增长[EB/OL].(2024-12-03)[2024-12-15].https://baijiahao.baidu.com/s?id=18173 85667802794294&wfr=spider&for=pc.
基本信息:
DOI：
中图分类号:TP391.3;G255.53
引用信息:
[1]张超然,梁素平,杨杰等.基于大模型RAG架构的专利检索Agent研究[J].中国发明与专利,2025,22(02):14-25.
基金信息: