Abstract: With the rapid growth of AI technologies, ensuring the security of open-source Large Language Models (LLMs) is crucial to maintaining their reliability and trustworthiness. The paper ...