Safeguarding Text-to-Image Generative Models Against Unauthorized Knowledge Distillation | ArxivCSExplorer