Open Access
- 28 June 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Software
- Vol. 41 (6) , 55-64
Abstract
High quality data is essential for designing effective software test suites. We propose three original methods for using large language models to generate representative test data, which fit to the domain of the program under test and are culturally adequate.Keywords
All Related Versions
This publication has 12 references indexed in Scilit:
- With Great Humor Comes Great Developer Engagement2024
- Breaking the Silence: the Threats of Using LLMs in Software Engineering2024
- An Empirical Evaluation of Using Large Language Models for Automated Unit Test GenerationIEEE Transactions on Software Engineering, 2023
- Enhancing Synthetic Test Data Generation with Language Models Using a More Expressive Domain-Specific Language2023
- Fill in the Blank: Context-aware Automated Text Input Generation for Mobile GUI Testing2023
- Application of Large Language Models to Software Engineering Tasks: Opportunities, Risks, and ImplicationsIEEE Software, 2023
- Can Pre-trained Language Models Understand Chinese Humor?2023
- MockSniffer2020
- Generating Fake Data Using GANs for Anonymizing Healthcare Data2020
- Synthetic Test Data Generation Using Recurrent Neural Networks: A Position Paper2019