Towards an Intelligent Test Case Generation Framework Using LLMs and Prompt Engineering

Software testing is a critical phase in software development, and the research for more intelligent and adaptive test case generation approaches is ever-growing. As software systems grow in complexity, the need for advanced methodologies becomes imperative and the integration of Natural Language Processing (NLP) and Large Language Models (LLMs) into software testing offers unparalleled advantages. In response to this demand, we present an innovative prototype for an intelligent test case generation framework that leverages the strengths of fine-tuning through the TuneReqTest module and user-centric collaboration via the PrompTester module. TuneReqTest refines a pre-trained LLM with Software Requirements Specification (SRS) documents and associated test cases, enabling contextual adaptability to software-specific nuances. PrompTester introduces an intuitive interface for testers to actively participate in the test case generation process, guiding the LLM through tailored prompts. Our framework excels in adaptability to evolving requirements, reduces dependency on predefined test cases, and enhances coverage and quality assurance. Through these advancements, our framework represents a significant leap forward in the landscape of automated software testing, fostering collaboration between advanced language models and human testers for more effective and context-aware test case generation.