Improving the Readability of Generated Tests Using GPT-4 and ChatGPT Code Interpreter
Paper in proceeding, 2024
A major challenge in automated test generation is the readability of generated tests. Emerging large language models (LLMs) excel at language analysis and transformation tasks. We propose that improving test readability is such a task and explore the capabilities of the GPT-4 LLM in improving readability of tests generated by the Pynguin search-based generation framework. Our initial results are promising. However, there are remaining research and technical challenges.
Large Language Models
Readability
Automated Test Generation
Search-Based Test Generation
Generative AI
Author
Gregory Gay
Software Engineering 1
University of Gothenburg
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
03029743 (ISSN) 16113349 (eISSN)
Vol. 14415 LNCS 140-1469783031487958 (ISBN)
15th International Symposium on Search-Based Software Engineering, SSBSE 2023
San Francisco, USA,
San Francisco, USA,
Context-Infused Automated Software Test Generation
Swedish Research Council (VR) (2019-05275), 2020-01-01 -- 2023-12-31.
Subject Categories (SSIF 2011)
Language Technology (Computational Linguistics)
Software Engineering
Computer Vision and Robotics (Autonomous Systems)
DOI
10.1007/978-3-031-48796-5_11