Seeing It Before It Happens: In-Generation NSFW Detection for Diffusion-Based Text-to-Image Models

Diffusion-based text-to-image (T2I) models enable high-quality image generation but also pose significant risks of misuse, particularly in producing not-safe-for-work (NSFW) content. While prior detection methods have focused on filtering prompts before generation or moderating images afterward, the in-generation phase of diffusion models remains largely unexplored for NSFW detection. In this paper, we introduce In-Generation Detection (IGD), a simple yet effective approach that leverages the predicted noise during the diffusion process as an internal signal to identify NSFW content. This approach is motivated by preliminary findings suggesting that the predicted noise may capture semantic cues that differentiate NSFW from benign prompts, even when the prompts are adversarially crafted. Experiments conducted on seven NSFW categories show that IGD achieves an average detection accuracy of 91.32% over naive and adversarial NSFW prompts, outperforming seven baseline methods.

Key Contributions

• Identifies predicted noise in diffusion denoising steps as a discriminative semantic signal that separates NSFW from SFW prompts, including adversarially obfuscated ones
• Proposes IGD (In-Generation Detection), a lightweight classifier trained on mid-generation noise features enabling early NSFW intervention before image synthesis completes
• Demonstrates IGD achieves 92.45% average detection accuracy across seven NSFW categories, outperforming seven baseline pre- and post-detection methods while remaining robust to adversarial prompts

🛡️ Threat Analysis

Details

Similar Papers