通过翻转注意力值，实现高效负向提示引导。
VSF: Simple, Efficient, and Effective Negative Guidance in Few-Step Image Generation Models By Value Sign Flip
- 翻转负向提示的注意力值，动态抑制不想要内容。
- 在少步生成模型中显著提升负向提示遵循度。
- 适配SD3.5 Turbo等架构，代码开源可用。
我们提出一种名为值符号翻转（Value Sign Flip, VSF）的方法，用于在少步扩散与流匹配图像生成模型中高效引入负向提示引导。与分类器自由引导（CFG）、NASA、NAG等方法不同，VSF通过翻转负向提示的注意力值实现对不良内容的动态抑制。该方法计算开销极小，可有效集成至MMDiT架构（如Stable Diffusion 3.5 Turbo）及基于交叉注意力的模型（如Wan）。我们在复杂提示对的挑战性数据集上验证了VSF，在静态图像与视频生成任务中均表现优异。实验表明，相较于以往方法，VSF在少步模型中显著提升负向提示遵循度，甚至优于非少步模型中的CFG，同时保持良好的图像质量。代码与ComfyUI节点已开源。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。