Modular Energy Steering for Safe Text-to-Image Generation with Foundation Models

Dimensions overall score 7.0

GitHub Code Pulse

No public code linked for this paper yet.

• The proposed inference-time steering framework uses gradient feedback from frozen pretrained foundation models to guide text-to-image generation without modifying the underlying generator.method 95%
• The method formulates safety steering as an energy-based sampling problem by injecting feedback through clean latent estimates at each sampling step.technical 90%
• The approach enables modular, training-free safety control that is compatible with both diffusion and flow-matching models.technical 90%
• The method demonstrates state-of-the-art robustness against NSFW red-teaming benchmarks.result 85%
• The method preserves high generation quality on benign non-targeted prompts while providing safety control.result 85%
• Existing safety approaches typically rely on model fine-tuning or curated datasets, which can degrade generation quality or limit scalability.limitation 80%
• Vision-language foundation models encode rich semantic representations that can be repurposed as off-the-shelf supervisory signals during generation.technical 85%
• The framework provides a principled approach for utilizing foundation models as semantic energy estimators for safety control.method 80%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card