Why Are My Prompts Leaked? Unraveling Prompt Extraction Threats in Customized Large Language Models

Open in new window