Research on AI-Empowered Software Test Case Generation
TL;DR
This work leverages large language models to analyze and extract test cases from requirement documents, and constructs an automated test case generation framework by specifying standard input and output values of test samples, test coverage scope, and AI agent roles.
Abstract
To resolve the problem that manually writing test cases required substantial effort and to promote software test automation, This paper explores the practice of applying AI technologies combined with large language models to test case generation within the software testing workflow. It leverages large language models to analyze and extract test cases from requirement documents, and constructs an automated test case generation framework by specifying standard input and output values of test samples, test coverage scope, and AI agent roles. Subsequent verification of generated test cases and iterative tuning of prompts and model parameters enable rapid generation of requirement-based test cases, delivering notable improvements in boundary testing and stress testing while shortening the test development cycle. This work expands application scenarios for the automated generation of software test cases and the engineering deployment of large language models.
