Back to Home
arXiv AI··Papers & Tech

Keep It CALM: Analyzing the Limits of Global Unsafety in Text-to-Image Generation

中文摘要

该研究分析了文生图全局安全信号的局限性,揭示了覆盖多样化不安全语义与避免扭曲安全内容之间的权衡。

English Summary

This study analyzes the limits of global safety signals in text-to-image generation, revealing a trade-off between covering diverse unsafe semantics and avoiding safe content distortion.

Original Excerpt

arXiv:2610.02300v1 Announce Type: new Abstract: 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: compact unsafe subspaces fail to cover heterogeneous unsafe semantics, whereas broader aggregation increasingly distorts safety-adjacent benign prompts. Motivated by this finding, we propose CALM (Counterfactual Adaptive Local Modulation), a training-free safeguard that replaces uniform global removal with prompt-local counterfactual correction. Using matched unsafe-benign anchors, CALM routes each prompt to active unsafe categories, minimally edits only violating token representations toward the safe side, and suppresses positively aligned unsafe residual components. Across broad evaluation, CALM significantly improves unsafe content suppression while preserving benign utility, demonstrating that local counterfactual correction provides a more selective alternative to global unsafe signal removal.