EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Automated Software Testing through Large Language Models: Opportunities and Challenges

Jayson A. Dela Fuente

Qubahan Techno Journal · 2022

DOI: 10.48161/qtj.v1n3a15 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Qubahan Techno Journal, 2022. Verfasst von Jayson A. Dela Fuente. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The rapid evolution of software systems and the increasing complexity of modern applications have driven the need for innovative testing methodologies. Traditional techniques, often burdened by manual intervention and limited scalability, are increasingly challenged by the demands for speed, flexibility, and comprehensive quality assurance. Large Language Models (LLMs) offer a promising avenue for transforming automated software testing. This research examines the opportunities and challenges of integrating LLMs into testing workflows, supported by extensive literature, empirical surveys, and real-world case studies.Key opportunities include automated test-case generation, program repair, test-oracle creation, debugging assistance, and the development of self-healing testing systems. For instance, surveys show that approximately 51% of studies focus on test-case generation, while 26% address program repair, 11% focus on test-oracle creation, and 12% on debugging assistance. Concurrently, several challenges hinder the seamless adoption of LLM-based testing. These include the risk of hallucinated outputs, context window limitations, security and privacy issues, high computational costs, integration complexities, and potential biases in generated tests.This paper comprehensively reviews the theoretical underpinnings and practical implementations of LLM-driven test automation while proposing pathways for future research and development. Ultimately, LLMs are positioned as transformative tools whose effective integration could greatly enhance testing efficiency and reliability. However, careful design and human oversight remain critical to overcome the inherent challenges.

Zitieren

APA

Fuente, J. A. D. (2022). Automated Software Testing through Large Language Models: Opportunities and Challenges. Qubahan Techno Journal. https://doi.org/10.48161/qtj.v1n3a15

Vancouver

Fuente JAD. Automated Software Testing through Large Language Models: Opportunities and Challenges. Qubahan Techno Journal. 2022. doi:10.48161/qtj.v1n3a15

BibTeX
@article{Fuente2022,
  author = {Fuente, Jayson A. Dela},
  title = {Automated Software Testing through Large Language Models: Opportunities and Challenges},
  journal = {Qubahan Techno Journal},
  year = {2022},
  doi = {10.48161/qtj.v1n3a15}
}
Kostenlos starten →