English(EN) Evaluating LLM-Based Goal Extraction in Requirements Engineering: Prompting Strategies and Their Limitations

LLM提示以61%的准确率提取软件目标，辅助人工工作

作者 PulseAugur 编辑部 · [2 个来源] · 2026-04-24 04:22

研究人员开发了一种使用链式LLM和工程化提示的方法，以自动从软件文档中提取功能目标。该方法包括参与者识别以及高/低级目标提取，并引入了生成-批评机制作为两个LLM之间的反馈循环。虽然该流程在低级目标识别方面达到了61%的准确率，但它最适合加速手动提取而非完全取代它。未来的工作旨在通过集成检索增强生成（RAG）和思维链（CoT）提示来提高准确性。 AI

影响提供了一个加速软件工程中手动目标提取的工具，并有可能通过RAG和CoT进行未来改进。

排序理由学术论文，详细介绍了基于LLM的软件工程目标提取的新方法。

在 arXiv cs.CL 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。我们如何撰写摘要 →

报道来源 [2]

arXiv cs.CL TIER_1 English(EN) · Anna Arnaudo, Riccardo Coppola, Maurizio Morisio, Flavio Giobergia, Andrea Bioddo, Angelo Bongiorno, Luca Dadone · 2026-04-27 04:00

Evaluating LLM-Based Goal Extraction in Requirements Engineering: Prompting Strategies and Their Limitations

arXiv:2604.22207v1 Announce Type: cross Abstract: Due to the textual and repetitive nature of many Requirements Engineering (RE) artefacts, Large Language Models (LLMs) have proven useful to automate their generation and processing. In this paper, we discuss a possible approach f…
arXiv cs.CL TIER_1 English(EN) · Luca Dadone · 2026-04-24 04:22

Evaluating LLM-Based Goal Extraction in Requirements Engineering: Prompting Strategies and Their Limitations

Due to the textual and repetitive nature of many Requirements Engineering (RE) artefacts, Large Language Models (LLMs) have proven useful to automate their generation and processing. In this paper, we discuss a possible approach for automating the Goal-Oriented Requirements Engin…

报道来源 [2]

Evaluating LLM-Based Goal Extraction in Requirements Engineering: Prompting Strategies and Their Limitations

Evaluating LLM-Based Goal Extraction in Requirements Engineering: Prompting Strategies and Their Limitations

相关实体

相关话题