Affiliation:
1. Riga Technical University , Riga , Latvia
Abstract
Abstract
Generative AI is only a few years old but already being applied in Software Engineering (SE). This literature review examines the most popular SE sub-fields of such cases and research methods that are typically used. 117 studies starting from 2020 have been assessed, and literature review has shown that the most active research is ongoing in the code generation area. It is not clearly defined by researchers, but the majority of the methods can be assumed as experiments. It is concluded that researchers often do not define the used research method with exclusions such as literature review or opinion survey. However, different validation methods are highly valued and applied thoroughly.
Reference26 articles.
1. G. Giray, “A software engineering perspective on engineering machine learning systems: State of the art and challenges,” Journal of Systems and Software, vol. 180, Oct. 2021, Art. no. 111031. https://doi.org/10.1016/j.jss.2021.111031
2. A. Mehra, M. Yadav, and B. Ammu, “GPT history.” [Online]. Available: https://www.kdnuggets.com/2023/05/deep-dive-gpt-models.html. Accessed on: Jan. 03, 2024.
3. A. Fan et al., “Large language models for software engineering: Survey and open problems,” arXiv.org, Oct. 2023. https://doi.org/10.48550/ARXIV.2310.03533
4. N. A. Parikh, “Empowering business transformation: The positive impact and ethical considerations of Generative AI in software product management – A systematic literature review,” arXiv.org, Jun. 2023. https://doi.org/10.48550/ARXIV.2306.04605
5. Z. Zheng et al., “Towards an understanding of large language models in software engineering tasks,” arXiv:2308.11396, Aug. 2023. https://doi.org/10.48550/arXiv.2308.11396