VSF: Simple, Efficient, and Effective Negative Guidance in Few-Step Image Generation Models By Value Sign Flip
This addresses the problem of controlling unwanted content in image generation for users of diffusion and flow-matching models, offering an incremental improvement over existing negative guidance techniques.
The paper tackles the problem of incorporating negative prompt guidance in few-step image generation models by introducing Value Sign Flip (VSF), a method that flips the sign of attention values from negative prompts to suppress undesired content. Experimental results show that VSF significantly improves negative prompt adherence compared to prior methods in few-step models and even CFG in non-few-step models, with superior performance on challenging datasets.
We introduce Value Sign Flip (VSF), a simple and efficient method for incorporating negative prompt guidance in few-step diffusion and flow-matching image generation models. Unlike existing approaches such as classifier-free guidance (CFG), NASA, and NAG, VSF dynamically suppresses undesired content by flipping the sign of attention values from negative prompts. Our method requires only small computational overhead and integrates effectively with MMDiT-style architectures such as Stable Diffusion 3.5 Turbo, as well as cross-attention-based models like Wan. We validate VSF on challenging datasets with complex prompt pairs and demonstrate superior performance in both static image and video generation tasks. Experimental results show that VSF significantly improves negative prompt adherence compared to prior methods in few-step models, and even CFG in non-few-step models, while maintaining competitive image quality. Code and ComfyUI node are available in https://github.com/weathon/VSF/tree/main.