让生成模型主动避开不良内容，提升图像质量与多样性
Signed Rectified Flow: Negativity-Controlled Generation
- 通过带符号的流模型控制正负分布，实现对生成区域的精准引导
- 在ImageNet上提升图像保真度与多样性，在抗记忆实验中降低相似性
- 适用于需排除特定内容的场景，如规避敏感生成或提升模型可控性
我们提出签名修正流（Signed Rectified Flow, Signed RF），其目标是生成一个带符号的测度 π^sign = (1+α)π^+ - απ^-，其中 α>0，π^+ 是希望生成的分布，π^- 是需要抑制的分布。尽管直接从带符号测度采样不被严格定义，但 Signed RF 构建了一个有效的生成过程：概率集中在符号测度为正的区域，同时可证明地排除负成分主导的区域。该方法为生成建模中融入负向信息与排除约束提供了理论依据。我们分析了支撑 Signed RF 的带符号连续性方程，并通过带电粒子类比解释负质量如何形成排斥屏障。这一理论进一步启发了自适应引导算法的设计。在多个应用中，Signed RF 在 ImageNet 上改善了保真度-多样性权衡，抗记忆实验中显著降低最近邻相似性，并在 Stable Diffusion 3.5 中有效减少对抗提示引发的裸露内容，同时保持 CLIP 与审美评分不变。
原文摘要 · Abstract (English)
We introduce Signed Rectified Flow (Signed RF), a generalization of Rectified Flow that targets the signed measure $π^{sign} = (1+α)π^+ - απ^-$, where $α>0$, $π^+$ is the distribution to promote, and $π^-$ is the distribution to suppress. Although direct sampling from a signed measure is not well-defined, Signed RF induces a valid generative process that concentrates probability in regions where the signed measure is positive while provably excluding regions dominated by its negative component. It therefore provides a principled framework for incorporating negative information and exclusion constraints into generative modeling. We analyze the signed continuity equation underlying Signed RF and use a charged-particle interpretation to explain how negative mass forms exclusion barriers. This theory further motivates practical adaptive guidance algorithms. Across several applications, Signed RF improves the fidelity-diversity trade-off on ImageNet, reduces nearest-neighbor similarity in anti-memorization experiments, and reduces nudity induced by adversarial prompts in Stable Diffusion 3.5 while preserving CLIP and aesthetic scores.
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。