用大模型从判例中自动提炼通用法律原则，提升司法推理能力。
Legal Rule Induction: Towards Generalizable Principle Discovery from Analogous Judicial Precedents
- 从相似判例中归纳共性要件、行为规范与法律后果，形成可泛化的法律规则。
- 在5121个案例集（38088个案件）上训练后，模型对规则模式的捕捉显著提升。
- 首个中文法律规则诱导基准，适合法律AI研究者和司法智能化开发者。
法律规则不仅包括成文法，还包含来自判例中蕴含的裁量性规范、社会道德与政策等隐含原则。尽管计算法学已在规则应用方面取得进展，但从司法判决中归纳法律规则的研究仍较薄弱。大语言模型（LLMs）为自动化提取这些隐含原则提供了前所未有的可能，但受限于任务定义不清与缺乏基准。为此，我们正式提出法律规则诱导（LRI）任务：从相关判例中提炼简洁、可泛化的教义规则，概括其共同前提、规范行为与法律后果。我们构建了可复现的LRI数据集生成流程，并以中国法为例，创建首个LRI基准，包含5,121个案例集（共38,088个法院案件）用于模型训练，以及216个专家标注的黄金测试集。实验表明：1）当前顶尖LLMs易出现过度泛化与幻觉；2）在本数据集上训练后，模型在捕捉相似案例间细微规则模式的能力显著增强。
原文摘要 · Abstract (English)
Legal rules encompass not only codified statutes but also implicit adjudicatory principles derived from precedents that contain discretionary norms, social morality, and policy. While computational legal research has advanced in applying established rules to cases, inducing legal rules from judicial decisions remains understudied across jurisdictions. The advent of Large Language Models (LLMs) offers unprecedented potential to automate the extraction of such latent principles, yet progress is stymied by the absence of formal task definitions and benchmarks. To address this gap, we formalize Legal Rule Induction (LRI) as the task of deriving concise, generalizable doctrinal rules from analogous precedents, distilling their shared preconditions, normative behaviors, and legal consequences. We further propose a reproducible pipeline for LRI dataset construction and, instantiating it on Chinese law as a representative jurisdiction, introduce the first LRI benchmark comprising 5,121 case sets (38,088 court cases) for model tuning and 216 expert-annotated gold test sets. Experimental results reveal that: 1) State-of-the-art LLMs struggle with over-generalization and hallucination; 2) Training on our dataset markedly enhances LLMs capabilities in capturing nuanced rule patterns across similar cases.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。