Parthenon Law: 一种自我进化的法律智能体框架
Parthenon Law: A Self-Evolving Legal-Agent Framework
AI总结：
本文提出Parthenon框架，通过分解模型、工具、知识等组件并引入反泄漏学习循环，使法律领域的大语言模型智能体能够从经验中自我进化，显著提升法律事务处理性能。
摘要：
随着智能体能力的增强，法律领域的大语言模型智能体有望将文档密集型事务转化为可审查的工作产品——然而可靠部署面临三个障碍：缺乏关于当前最强模型与框架组合在端到端法律事务上行为的大规模证据；没有适应法律垂直领域的智能体架构，只有通用框架；以及在不断变化的事实、权威和截止日期环境中，缺乏系统从自身结果中学习的机制。我们逐一解决这些问题。在Harvey LAB上进行的大规模实证研究——包含12,510条智能体轨迹——表明即使是前沿智能体也无法一次性完成事务：每项标准的准确率随模型增强而提高，但严格的事务完成率停滞不前。然后我们引入Parthenon，一种自我进化的法律智能体框架，将模型、框架、智能体角色、法律知识、确定性工具和程序技能分解为可审计的表面，以实现来源可追溯性、日期和数字接地、交付物合规性和问题关闭。最后，一个反泄漏学习循环将评分失败转化为对技能、工具和知识的任务无关编辑，使系统能够随着经验改进——就像律所在每个事务后完善其检查清单和操作手册——而不触及模型权重。在我们的大规模实证分析中，Parthenon显著提升了最先进模型和框架在法律事务任务上的性能。
英文摘要：
As agents grow more capable, legal-domain LLM agents promise to turn document-heavy matters into reviewable work products -- yet reliable deployment faces three obstacles: no large-scale evidence on how today's strongest model-and-harness combinations behave on end-to-end legal matters; no agent architecture adapted to the legal vertical, only general-purpose harnesses; and, in a setting that keeps shifting with new facts, authorities, and deadlines, no mechanism for systems to learn from their own outcomes. We address each. A large-scale empirical study on Harvey LAB -- $12{,}510$ agent trajectories -- shows that even frontier agents remain far from completing matters in a single pass: per-criterion accuracy climbs with stronger models while strict matter completion stalls. We then introduce \textsc{Parthenon}, a self-evolving legal-agent framework that factors Model, Harness, Agent roles, legal Knowledge, deterministic Tools, and procedural Skills into auditable surfaces for source traceability, date and number grounding, deliverable compliance, and issue closure. Finally, an anti-leakage learning loop converts scored failures into task-agnostic edits to skills, tools, and knowledge, letting the system improve with experience -- as a firm refines its checklists and playbooks after each matter -- without touching model weights. Across our large-scale empirical analysis, \textsc{Parthenon} substantially improves the performance of state-of-the-art models and harnesses on legal-matter tasks.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
