Quality Assurance for LLM-Generated Test Cases: A Systematic Literature Review

The rapid advancements in artificial intelligence have transformed software testing, with Large Language Models (LLMs) emerging as powerful tools for automating test case generation. This paper explores Quality Assurance (QA) for LLM-generated test cases in black-box testing through a systematic literature review. Though LLMs are increasingly used for test case generation, challenges in ensuring their quality remain. Following PRISMA guidelines, relevant studies were selected from databases focusing on critical quality attributes, QA frameworks, metrics, and challenges. LLMs demonstrate high efficiency but face numerous issues. A recommendation for future research is given on addressing standardized metrics and improving human-AI collaboration for enhanced testing outcomes.