Published: Oct. 18, 2024
Language: Английский
Empirical Software Engineering, Journal Year: 2024, Volume and Issue: 30(2)
Published: Dec. 26, 2024
Language: Английский
Citations
10
Cluster Computing, Journal Year: 2024, Volume and Issue: 27(8), P. 10717 - 10731
Published: May 9, 2024
Language: Английский
Citations
4
Software, Journal Year: 2025, Volume and Issue: 4(1), P. 5 - 5
Published: March 10, 2025
Software testing ensures the quality and reliability of software products, but manual test case creation is labor-intensive. With rise Large Language Models (LLMs), there growing interest in unit with LLMs. However, effective assessment LLM-generated cases limited by lack standardized benchmarks that comprehensively cover diverse programming scenarios. To address an LLM’s generation ability lacking a dataset for evaluation, we propose Generated Benchmark from Control-Flow Structure Variable Usage Composition (GBCV) approach, which systematically generates programs used evaluating LLMs’ capabilities. By leveraging basic control-flow structures variable usage, GBCV provides flexible framework to create spectrum ranging simple complex. Because GPT-4o GPT-3.5-Turbo are publicly accessible models, present real-world regular users’ use cases, assess LLM performance on them. Our findings indicate performs better composite program structures, while all models effectively detect boundary values conditions face challenges arithmetic computations. This study highlights strengths limitations LLMs generation, benchmark framework, suggests directions future improvement.
Language: Английский
Citations
0
Computers in Human Behavior Artificial Humans, Journal Year: 2025, Volume and Issue: unknown, P. 100144 - 100144
Published: March 1, 2025
Language: Английский
Citations
0
ACM Transactions on Software Engineering and Methodology, Journal Year: 2025, Volume and Issue: unknown
Published: Jan. 27, 2025
The effectiveness of a test suite in detecting faults highly depends on the quality its oracles. Large Language Models (LLMs) have demonstrated remarkable proficiency tackling diverse software testing tasks. This paper aims to present roadmap for future research use LLMs oracle automation. We discuss progress made field automation before introduction LLMs, identifying main limitations and weaknesses existing techniques. Additionally, we recent studies this task, highlighting challenges that arise from their use, e.g., how assess usefulness generated conclude with discussion about directions opportunities LLM-based
Language: Английский
Citations
0
ACM Transactions on Software Engineering and Methodology, Journal Year: 2025, Volume and Issue: unknown
Published: April 25, 2025
The landscape of software engineering has dramatically changed in recent years. impressive advances artificial intelligence are just the latest and most disruptive innovation that remarkably research practice. This special issue shares a roadmap to guide community this confused era. is outcome two-day intensive discussion at 2030 Software Engineering workshop. spotlights discusses seven main landmarks new landscape: engineering, human aspects security, verification validation, sustainable quantum engineering. editorial summarizes core discussed 37 papers comprise sections guides interested readers throughout issue. living body we will refine with follow-up workshops update for series forthcoming ACM TOSEM issues.
Language: Английский
Citations
0
Empirical Software Engineering, Journal Year: 2025, Volume and Issue: 30(3)
Published: April 26, 2025
Language: Английский
Citations
0
Published: Jan. 1, 2024
Software engineering researchers and practitioners have pursued manners to reduce the amount of time effort required develop code increase productivity since emergence discipline. Generative language models are another step in this journey, but it will probably not be last one. In chapter, we propose DAnTE, a Degree Automation Taxonomy for software Engineering, describing several levels automation based on idiosyncrasies field. Based taxonomy, evaluated tools used past present practices. Then, give particular attention AI-based tools, including generative models, discussing how they located within proposed taxonomy reasoning about possible limitations currently have. analysis, discuss novel that could emerge middle long term.
Language: Английский
Citations
1
Published: Oct. 18, 2024
Language: Английский
Citations
1
ACM Transactions on Software Engineering and Methodology, Journal Year: 2024, Volume and Issue: unknown
Published: Dec. 18, 2024
Large Language Models (LLMs) have become instrumental in advancing software engineering (SE) tasks, showcasing their efficacy code understanding and beyond. AI models has demonstrated value not only generating but also defect detection, enhancing security measures, improving overall quality. They are emerging as crucial tools for both development maintaining Like traditional SE tools, open-source collaboration is key realising the excellent products. However, with models, essential need data. The of these AI-based hinges on maximising sources high-quality data especially high quality, often holds commercial or sensitive value, making it less accessible projects. This reality presents a significant barrier to enhancement within community. Therefore, researchers find solutions enabling tap into resources by different organizations. Addressing this challenge, our position paper investigates one solution facilitate access diverse organizational ensuring privacy sensitivities respected. We introduce governance framework centered federated learning (FL), designed foster joint maintenance while safeguarding security. Additionally, we present guidelines developers tool collaboration, covering requirements, model architecture, updating strategies, version control. Given influence characteristics learning, research examines effect heterogeneity performance. consider 6 scenarios distributions include 4 models. most common algorithms. Our experimental findings highlight potential employing collaborative discuss issues be addressed co-construction process future directions.
Language: Английский
Citations
1