计算机科学

计算机科学 ›› 2026, Vol. 53 ›› Issue (7): 118-124.doi: 10.11896/jsjkx.250600009

• 人工智能 • 上一篇    下一篇

从布尔检索到基座模型:类案检索技术发展的法学审视

石依冉1, 张凌寒2, 刘奕群3   

  1. 1 中国政法大学数据法治研究院 北京 100088
    2 中国政法大学人工智能法研究院 北京 100088
    3 清华大学计算机科学与技术系 北京 100084
  • 收稿日期:2025-06-03 修回日期:2025-11-21 出版日期:2026-07-15 发布日期:2026-07-10
  • 通讯作者: 张凌寒(zl343@cupl.edu.cn)
  • 作者简介:(2402420299@cupl.edu.cn)
  • 基金资助:
    2024年度司法部法治建设与法学理论研究重点课题(24SFB1011);2025年中国政法大学科研创新项目(10825383)

From Boolean Retrieval to Foundational Models:Legal Perspective on Development of Case Retrieval Technology

SHI Yiran1, ZHANG Linghan2, LIU Yiqun3   

  1. 1 The Institute for Data Law,China University of Political Science and Law,Beijing 100088,China
    2 The Institute of AI Law and Governance,China University of Political Science and Law,Beijing 100088,China
    3 Department of Computer Science and Technology,Tsinghua University,Beijing 100084,China
  • Received:2025-06-03 Revised:2025-11-21 Published:2026-07-15 Online:2026-07-10
  • About author:SHI Yiran,born in 1999,Ph.D canidate.Her main research interests include data law and AI law.
    ZHANG Linghan,born in 1982,Ph.D,professor,Ph.D supervisor.Her main research interests include AI(algorithms),data,and platform governance.
  • Supported by:
    2024 Ministry of Justice Key Research Project on Construction of Rule of Law and Legal Theory(24SFB1011) and 2025 China University of Political Science and Law Research Innovation Project(10825383).

PDF (PC)

20

摘要： 以人工智能为代表的技术创新正深刻影响司法实践,推动“智慧法院”建设进程。在这一过程中,类案检索技术作为人工智能应用的重要环节,正从传统的依赖关键词匹配和浅层文本特征的检索模式,逐步向基于深层语义理解、语境感知与逻辑推理的智能检索转型。人工智能加持下的类案检索技术能够在一定程度上模拟法官在审理案件时的裁判逻辑,成为提升检索效率、推动法律适用标准统一、促进司法公正的重要手段。然而,对实务一线法官的深入访谈显示,当前司法实践中应用的类案检索技术与科研实验中的技术水平之间仍存在较大差距,主要体现在3个方面:1)检索方式仍以关键词检索为主,智能语义检索尚未广泛普及,检索效率较低;2)检索质量不足,现有平台数据库在案例数量与案例权威性之间难以平衡,且缺乏统一、明晰的“类案”判断标准;3)检索规则不完善,缺乏对案件地域、审级、裁判效力等因素的精细划分。因此,以现阶段主流类案检索平台的应用情况为基础,采用横向平台对比与纵向技术演进分析相结合的方法,从法学理论与司法实务的双重视角出发,探讨类案检索技术发展现状及其存在的瓶颈问题,并在此基础上提出未来类案检索技术与应用的改进方向。

关键词: 类案检索, 预训练模型, 法律, 人工智能, 智慧法院

Abstract: Technological innovation,exemplified by artificial intelligence,is reshaping judicial practice and driving the development of “smart courts”.In this context,similar case retrieval technology,as a key application of AI,is evolving from traditional keyword-based searches and shallow text matching to intelligent retrieval based on deep semantic understanding,contextual awareness,and logical reasoning.AI-powered retrieval can partially simulate judges' reasoning processes,improving efficiency,promoting consistency in legal application,and enhancing judicial fairness.However,in-depth interviews with frontline judges reveal a significant gap between the technology used in practice and the advancements seen in research.This gap manifests in three main respects:1)Etrieval methods remain predominantly keyword-based,with semantic-level retrieval yet to be widely adopted,resulting in limited efficiency.2)Retrieval quality remains inadequate,as existing databases struggle to balance case volume with case authority,and a unified and explicit standard for defining “similar cases” is still lacking.3)Retrieval rules are incomplete,with insufficient differentiation based on jurisdiction,trial level,and the legal effect of judgments.Drawing on the application of mainstream retrieval platforms,this paper adopts a mixed method combining cross-platform comparison with longitudinal technical analysis.From both legal theory and judicial practice perspectives,it examines the current development and existing bottlenecks of similar case retrieval technology and proposes directions for future improvement.

Key words: Legal case retrieval, Pre-trained model, Law, Artificial Intelligence(AI), Smart courts

中图分类号: 

  • TP181
[1]ZHANG L H.Concerns and Countermeasures of Technology Dependence in Smart Justice [J].Law and Social Development,2022,28(4):180-200.
[2]最高人民法院.关于加快建设智慧法院的意见[EB/OL].(2017-04-12) [2025-05-21].http://gongbao.court.gov.cn/Details/5dec527431cdc22b72163b49fc0284.html.
[3]最高人民法院.关于加快建设智慧法院的意见[EB/OL].(2017-04-12) [2025-05-21].http://gongbao.court.gov.cn/Details/5dec527431cdc22b72163b49fc0284.html.
[4]最高人民法院.关于完善统一法律适用标准工作机制的意见[EB/OL].(2020-09-23) [2025-05-21].https://www.court.gov.cn/fabu/xiangqing/258001.html.
[5]最高人民法院.关于规范和加强人工智能司法应用的意见[EB/OL].(2022-12-09) [2025-05-21].https://www.chinacourt.org/law/detail/2022/12/id/150461.shtml.
[6]SUN H B.How Does Case Retrieval Contribute to Consistent Adjudication?[J].Tsinghua Law Journal,2021,15(1):79-97.
[7]张昊.最高法发布“法信法律基座大模型”[N].法治日报,2024-11-16(003).
[8]HUTCHINSON B,ROSTAMZADEH N,GREER C,et al.Evaluation gaps in machine learning practice [C]//Proceedings of the 2022 ACM Conference on Fairness,Accountability,and Transparency(FAccT '22).ACM,2022:1859-1876.
[9]中国审判.十年“智”变[EB/OL].(2022-08-03) [2025-05-21].https://www.thepaper.cn/newsDetail_forward_19288750.
[10]人民法院报.数字法院的实践与展望[EB/OL].(2024-11-21) [2026-05-10] https://www.chinacourt.cn/article/detail/2024/11/id/8215844.shtml.
[11]CHEN K.Embedding Case Recommendation into “Smart Court” Trial Scenarios:Principles and Approaches [J].Chinese Journal of Applied Jurisprudence,2018(4):88-97.
[12]ZHOU J H,SI Y L,YU H S,et al.Key Issues in the Construction of the People's Court Case Database [J].Chinese Journal of Applied Jurisprudence,2024(3):16-28.
[13]ZUO W M.How to Achieve Consistent Adjudication via Artificial Intelligence [J].China Law Review,2018(2):26-32.
[14]GAO S.Standards for Determining Similar Legal Cases and Their Application [J].Legal Science(Journal of Northwest University of Political Science and Law),2020,38(1):24-35.
[15]YU T Z.On the Referenced Application of Guiding Cases [J].People's Justice,2013(7):62-66.
[16]LI H,SHAO Y,WU Y,et al.LeCaRDv2:A Large-Scale Chinese Legal Case Retrieval Dataset [C]//Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval(SIGIR '24).ACM,2024:2251-2260.
[17]FEI Z,SHEN X,ZHU D,et al.Law Bench:Benchmarking Legal Knowledge of Large Language Models [C]//Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing(EMNLP 2024).Association for Computational Linguistics,2024:7933-7962.
[18]ZHOU S H.Treating Like Cases Alike:A Fictional Myth of the Rule of Law [J].Legal Science,2015(11):131-140.
[19]SUN H B.Treating Like Cases Alike:Not a Fictional Myth of the Rule of Law [J].The Jurist,2019(5):141-157,195-196.
[20]MASON D.Legal information retrieval study-Lexis Professio-nal and Westlaw UK [J].Legal Information Management,2006,6(4):246-250.
[21]TURTLE H.Text retrieval in the legal world [J].Artificial Intelligence and Law,1995,3(1/2):15-54.
[22]KUMAR S,REDDY P K,REDDY V B,et al.Similarity analysis of legal judgments[C]//Proceedings of the Fourth Annual ACM Bangalore Conference.2011:1-4.
[23]SALTON G,BUCKLEY C.Term-weighting approaches in automatic text retrieval[J].Information processing & management,1988,24(5):513-523.
[24]TAMSIN M,BURKHARD S.Concept and context in legal information retrieval [C]//Proceedings of the 21st Conference on Legal Knowledge and Information Systems.IOS Press,2008:63-72.
[25]CLAUDIO C,GIOVANNI R.A survey of automatic query expansion in information retrieval [J].ACM Computing Surveys,2012,44(1):1-50.
[26]LEI B S.Methodology of Case Similarity Judgment [J].Modern Law Science,2022,44(6):24-41.
[27]LI Q Q,CHEN Y X,ZHANG X T,et al.MUSER:A multi-view similar case retrieval dataset [C]//Proceedings of the 32nd ACM International Conference on Information and Knowledge Management(CIKM '23).ACM,2023:5336-5340.
[28]KULKARNI Y H,PATIL R,SHRIDHARAN S.Detection of Catchphrases and Precedence in Legal Documents[C]//FIRE(Working Notes).2017:86-89.
[29]SHAO Y,LIU X,WU H,et al.BERT-PLI:Modeling paragraph-level interactions for legal case retrieval [C]//Proceedings of the 29th International Joint Conference on Artificial Intelligence(IJCAI '20).IJCAI,2020:3501-3507.
[30]DENG J T.Computability Research on Case Similarity Stan-dards in Consistent Adjudication [J].Law and Social Development,2024,30(5):205-224.
[31]HONG Z,ZHOU Q,ZHANG R,et al.Legal feature enhanced semantic matching network for similar case matching[C]//2020 International Joint Conference on Neural Networks(IJCNN).IEEE,2020:1-8.
[32]RABELO J,KIM M Y,GOEBEL R.Semantic-based classifica-tion of relevant case law[C]//JSAI International Symposium on Artificial Intelligence.Cham:Springer,2022:84-95.
[33]MA S,CHEN C,CHU Q,et al.Leveraging large language mo-dels for relevance judgments in legal case retrieval[J].arXiv:2403.18405,2024.
[34]STATE V.Loomis,Wisconsin Supreme Court Requires Warning Before Use of Algorithmic Risk Assessments in Sentencing [J].Harvard Law Review,2017,130:1530-1537.
[35]ASAI A,WU Z,WANG Y,et al.Self-rag:Learning to retrieve,generate,and critique through self-reflection[C]//The Twelfth International Conference on Learning Representations.2023.
[36]HOU A B,WELLER O,QIN G,et al.CLERC:A Dataset for Legal Case Retrieval and Retrieval-Augmented Analysis Gene-ration[J].arXiv:2406.17186,2024.
[37]SHU D,ZHAO H,LIU X,et al.LawLLM:Law large language model for the US legal system[C]//Proceedings of the 33rd ACM International Conference on Information and Knowledge Management.2024:4882-4889.
[38]最高人民法院.关于规范和加强人工智能司法应用的意见[EB/OL].(2022-12-09) [2025-05-21].https://www.chinacourt.org/law/detail/2022/12/id/150461.shtml.
[39]SUN H B.Rethinking the Possibilities and Limitations of Intelligent Adjudication [J].Journal of the National Prosecutors College,2020,28(5):80-99.
[40]GU P D.The Function Orientation of Judicial Precedents with Different Natures under China's Statutory Law System [J].China Legal Science,2021(4):5-24.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市两江新区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发