Systematic literature review identifies key applications of large language models in requirements engineering, indicating significant evolution in methodology and scope.
Key Points
The study reveals LLMs primarily aid in requirements elicitation and validation, shifting from past applications focused on defect detection.
A total of 74 primary studies published from 2023 to 2024 were systematically reviewed, demonstrating notable trends in research directions and challenges.
Methodological assessments highlight the predominant use of GPT-based models and prompting strategies like Zero-shot and Few-shot within controlled environments.
Future directions are outlined, emphasizing the need for real-world testing, expanded application scopes, and improved integration of LLMs in software engineering.