Safety-Guided Flow (SGF): A Unified Framework for Negative Guidance in Safe Generation

TL;DR

This paper introduces a unified probabilistic framework for safe image generation using negative guidance, combining control barrier functions and data-driven approaches to improve safety and quality.

Contribution

It presents a novel energy-based negative guidance method that unifies existing safety techniques and identifies a critical time window for guidance application.

Findings

Negative guidance is most effective in early denoising stages.

The framework unifies control barrier functions with data-driven safety methods.

Applying guidance outside the critical window can compromise safety and quality.

Abstract

Safety mechanisms for diffusion and flow models have recently been developed along two distinct paths. In robot planning, control barrier functions are employed to guide generative trajectories away from obstacles at every denoising step by explicitly imposing geometric constraints. In parallel, recent data-driven, negative guidance approaches have been shown to suppress harmful content and promote diversity in generated samples. However, they rely on heuristics without clearly stating when safety guidance is actually necessary. In this paper, we first introduce a unified probabilistic framework using a Maximum Mean Discrepancy (MMD) potential for image generation tasks that recasts both Shielded Diffusion and Safe Denoiser as instances of our energy-based negative guidance against unsafe data samples. Furthermore, we leverage control-barrier functions analysis to justify the existence of a critical time window in which negative guidance must be strong; outside of this window, the guidance should decay to zero to ensure safe and high-quality generation. We evaluate our unified framework on several realistic safe generation scenarios, confirming that negative guidance should be applied in the early stages of the denoising process for successful safe generation.