Large Language Models are Few-shot Testers: Exploring LLM-based General Bug Reproduction | IEEE Conference Publication | IEEE Xplore