Search-Based Software Engineering in the Era of Modern Software Systems

Sarro, Federica

doi:10.1109/re57278.2023.00010

Cited by 7 publications

(1 citation statement)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, methods for automating software maintenance and optimisation have been proposed. Although a lot of attention has been devoted to automated bug fixing, more and more techniques for improvement of non-functional properties of software such as runtime or memory consumption have emerged [2][3][4].…”

Section: Introductionmentioning

confidence: 99%

Large Language Model Based Mutations in Genetic Improvement

Brownlee,

Callan,

Even-Mendoza

et al. 2024

Preprint

View full text Add to dashboard Cite

Ever since the first large language models (LLMs) have become available, both academics and practitioners have used them to aid software engineering tasks. However, little research as yet has been done in combining search-based software engineering (SBSE) and LLMs. In this paper, we evaluate the use of LLMs as mutation operators for genetic improvement (GI), an SBSE approach, to improve the GI search process. In a preliminary work, we explored the feasibility of combining the Gin Java GI toolkit with OpenAI LLMs in order to generate an edit for the JCodec tool. Here we extend this investigation involving three LLMs and three types of prompt, and five real-world software projects. We sample the edits at random, as well as using local search. Our results show that, compared with conventional statement GI edits, LLMs produce fewer unique edits, but these compile and pass tests more often, with the OpenAI model finding test-passing edits 77% of the time. The OpenAI and Mistral LLMs were roughly equal in finding the best run-time improvements. Simpler prompts were more successful than those providing more context and examples. Qualitative analysis revealed a wide variety of areas where LLMs typically failed to produce valid edits: commonly including inconsistent formatting, generating non-Java syntax, or refusing to provide a solution.

show abstract

Section: Introductionmentioning

confidence: 99%

Large Language Model Based Mutations in Genetic Improvement

Brownlee,

Callan,

Even-Mendoza

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

Developer Views on Software Carbon Footprint and Its Potential for Automated Reduction

Lyu,

Gay,

Sakamoto

2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Reducing software carbon footprint could contribute to efforts to avert climate change. Past research indicates that developers lack knowledge on energy consumption and carbon footprint, and existing reduction guidelines are difficult to apply. Therefore, we propose that automated reduction methods should be explored, e.g., through genetic improvement. However, such tools must be voluntarily adopted and regularly used to have an impact. In this study, we have conducted interviews and a survey (a) to explore developers' existing opinions, knowledge, and practices with regard to carbon footprint and energy consumption, and (b), to identify the requirements that automated reduction tools must meet to ensure adoption. Our findings offer a foundation for future research on practices, guidelines, and automated tools that address software carbon footprint.

show abstract