paperbot · PL 论文追踪

全部论文 待读 收藏

P2 RE-oriented Model Development with LLM Support and Deduction-based Verification

面向需求工程的模型开发:LLM 支持与基于演绎的验证
FSE Proceedings of the 33rd ACM International Conference on the Foundations of Software Engineering2025匹配度 3/10 · 论文属于 LLM 辅助需求工程与 UML 行为模型向形式规约的自动翻译,仅在“LLM 与演绎验证结合”这一大方向上与画像的 AI→PL 分支有弱关联;其验证对象是需求/设计层的逻辑规约,而非代码级静态分析,也未涉及 Agentic/ML 程序、浮点或贝叶斯程序分析。引用 3 inbox
Radoslaw Klimek
crossref: 2026-09-12

AI 速览

一句话:提出一个结合 UML 建模、LLM 辅助与演绎验证的需求工程框架,可自动生成程序骨架。

需求工程对软件质量至关重要,但将建模、LLM 与形式验证整合仍具挑战。本文提出一个面向需求工程的综合框架,聚焦特定 UML 图以支持初步系统开发。框架在各建模阶段提供可视化,并无缝集成大语言模型与逻辑推理引擎:由 LLM 辅助生成的行为模型可自动转换为形式逻辑规约,再通过演绎形式验证检查逻辑需求及软件制品间关系是否被充分处理。最终,框架支持自动生成程序骨架,简化从设计到实现的过渡。摘要未报告实验评估,贡献主要在框架设计与流程集成。

问题
如何将需求工程中的建模、LLM 辅助生成和形式验证整合,以提升初步系统开发并顺畅过渡到实现。
方法
提出面向特定 UML 图的综合框架,在建模各阶段提供可视化,集成 LLM 与逻辑推理引擎,将 LLM 辅助生成的行为模型自动翻译为形式逻辑规约,并进行演绎验证,最终生成程序骨架。
评估
摘要未给出实证评估;验证机制体现为用演绎形式验证检查逻辑需求及软件制品间关系,并以自动生成程序骨架展示流程贯通。
原文摘要(Abstract)

The requirements engineering (RE) phase is pivotal in developing high-quality software. Integrating advanced modelling techniques with large language models (LLMs) and formal verification in a logical style can significantly enhance this process. We propose a comprehensive framework that focuses on specific Unified Modelling Language (UML) diagrams for preliminary system development. This framework offers visualisations at various modelling stages and seamlessly integrates large language models and logical reasoning engines. The behavioural models generated with the assistance of LLMs are automatically translated into formal logical specifications. Deductive formal verification ensures that logical requirements and interrelations between software artefacts are thoroughly addressed. Ultimately, the framework facilitates the automatic generation of program skeletons, streamlining the transition from design to implementation.

链接与引用

DOI 原文 · PDF(开放获取) ·

BibTeX
@inproceedings{paperbot9036,
  title = {RE-oriented Model Development with LLM Support and Deduction-based Verification},
  author = {Radoslaw Klimek},
  booktitle = {FSE},
  year = {2025},
  doi = {10.1145/3696630.3730562}
}