基于大小模型协同的层次化法律罪名判定方法、设备、介质

By employing a hierarchical legal crime determination method that combines the advantages of both large and small models, the problem of insufficient accuracy in legal crime determination in existing technologies is solved, achieving efficient crime determination.

CN120745802BActive Publication Date: 2026-07-17ZHEJIANG UNIV
View PDF 1 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
ZHEJIANG UNIV
Filing Date
2025-06-17
Publication Date
2026-07-17

Smart Images

  • Figure CN120745802B_ABST
    Figure CN120745802B_ABST
Patent Text Reader

Abstract

本发明公开了一种基于大小模型协同的层次化法律罪名判定方法、设备、介质，包括：首先，针对每大模型获取候选罪名集合：构建层次化罪名结构体系；大模型响应目标案件输出最底层罪名列表(第一列表)；将底层罪名与父罪名拼接后输出新列表(第二列表)；基于目标案件和K个相似案件及其罪名输出列表(第三列表)；基于目标案件和K个相似法条及其罪名输出列表(第四列表)；基于目标案件、相似案件和法条输出列表(第五列表)；取所有列表的并集作为该大模型候选集合。其次，合并所有大模型候选集合为最终候选集合；输入至小模型，小模型输出各罪名概率，取最大概率值对应罪名为最终罪名。
Need to check novelty before this filing date? Find Prior Art