一种基于结构化提示词与分节生成的规划文本智能生成方法

By constructing a planning text generation method based on structured prompt words and logical state transition networks, the problem of unclear logical state transmission between chapters in planning texts is solved, achieving logical consistency and semantic coherence of planning texts, and improving the reliability and applicability of the generated texts.

CN121743484BActive Publication Date: 2026-07-17GUANGXI TEACHERS EDUCATION UNIV +1
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
GUANGXI TEACHERS EDUCATION UNIV
Filing Date
2025-11-06
Publication Date
2026-07-17

Smart Images

  • Figure CN121743484B_ABST
    Figure CN121743484B_ABST
Patent Text Reader

Abstract

本发明涉及文本生成技术领域，具体涉及一种基于结构化提示词与分节生成的规划文本智能生成方法，包括S1：接收用户初始规划意图，识别并抽取出规划语义单元；S2：基于逻辑状态属性构建一个逻辑状态转移网；依据所述逻辑状态转移网，为每个待生成的规划章节生成包括前序状态约束的动态结构化提示词；S3：将动态结构化提示词输入语言模型，依次生成各章节内容；每生成一节，逆向提取出新的规划语义单元，将新规划语义单元的逻辑状态属性与逻辑状态转移网的预期目标进行比对，若不一致，则触发内容重构，最终输出逻辑状态一致的完整规划文本。本发明提升了规划文本生成的一致性、结构性与合理性，增强了应用实用性与适配广度。
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Intelligent writing method based on multi-agent cooperation

    CN120632092A

  • Generating a document structure

    US20150339278A1