arXiv cs.CL· Xinyu Yang, Jiejun Tan, Chenlong Deng, Chenghao Zhang, Tongyu Wen, Zhicheng Dou·· 11 小时前AI 评分29
LegalEvolver：基于评分标准引导自我进化的智能体法律判决预测
LegalEvolver: Agentic Legal Judgment Prediction via Rubric-Guided Self-Evolution
AI 导读
LegalEvolver 是一个面向法律判决预测（LJP）检索的查询优化框架，通过启发式优化查询作为种子提升搜索轨迹生成效率，并将 rollout 经验总结为评分标准（rubrics）引导多轮迭代自我进化。该方法迭代进行评分标准优化与评分标准引导的 RL，使模型突破初始性能上限。多基准端到端实验显示查询优化对 LJP 有效，且进化评分标准优于固定评分标准。
正文 · 原文
Abstract:As LLM-based agents develop, they have been increasingly applied in the legal domain. Legal Judgment Prediction~(LJP) requires models to conduct multi-step reasoning and external legal evidence retrieval. Existing legal search agents' performance is often constrained by the quality of retrieval queries. They often formulate queries directly from the initial user request, overlooking latent legal issues and legally salient details. To address this issue, we propose \emph{LegalEvolver}, a query optimization framework specifically designed to generate legally informative queries for LJP-oriented retrieval. Sampling high-quality optimized queries and their corresponding search trajectories is challenging, as relying solely on unguided high-temperature rollouts is both inefficient and constrained by the model's initial capabilities. We therefore heuristically optimize queries as seeds, improving the efficiency of trajectory generation. Furthermore, we summarize rollout experience into rubrics and use them to guide multi-round iterative self-evolution. Iteratively performing rubric optimization and rubric-guided RL enables the model to surpass its initial performance ceiling. End-to-end performance on extensive benchmarks demonstrates the effectiveness of query optimization on LJP. Comparisons with direct SFT followed by RL demonstrate the effectiveness of rubric guidance. Evolving rubrics further outperform fixed rubrics, confirming the benefits of self-evolution.
| Comments: | 20 pages, 5 figures |
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL) |
| Cite as: | arXiv:2611.06381 [cs.AI] |
| (or arXiv:2611.06381v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.06381 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Xinyu Yang [view email]
[v1]
Wed, 4 Nov 2026 11:45:46 UTC (3,680 KB)
来源：arXiv cs.CL · arxiv.org