Keep It CALM: Analyzing the Limits of Global Unsafety in Text-to-Image Generation
Training-free safeguards for text-to-image generation often rely on a reusable safety signal, such as an unsafe direction or global toxic subspace, applied broadly across prompts. We provide a controlled geometric analysis of this global-unsafety assumption and reveal a consistent coverage-selectivity trade-off: compac...
Nahyeon Park, Minhyun Lee, Hyunjung Shim
· 0 citations