通过语义退化条件引导扩散模型
Guiding Diffusion Models with Semantically Degraded Conditions
AI总结：
通过语义退化条件引导扩散模型，提升文本图像对齐与组合准确性。
摘要：
分类器自由引导（CFG）是现代文本到图像模型的核心，但其依赖于语义空的空提示（∅）生成易受几何纠缠的引导信号。这是限制其精度的关键因素，导致在复杂组合任务中出现已知的失败。我们提出条件退化引导（CDG），一种新的范式，用战略性退化的条件替换空提示，c_deg。这将引导从粗略的“好 vs. 空”对比转变为更精细的“好 vs. 几乎好”辨别，从而迫使模型捕捉细粒度的语义区别。我们发现transformer文本编码器中的token分为两个功能角色：内容token编码对象语义，和上下文聚合token捕捉全局上下文。通过仅选择性地退化前者，CDG构建c_deg而无需外部模型或训练。在多样化的架构中验证，包括Stable Diffusion 3、FLUX和Qwen-Image，CDG显著提高了组合准确性和文本图像对齐。作为轻量级、即插即用的模块，它在计算开销极小的情况下实现了这一点。我们的工作挑战了对静态、信息稀疏的负样本的依赖，并建立了扩散引导的新原则：构建适应性、语义感知的负样本对于实现精确的语义控制至关重要。代码可在https://github.com/Ming-321/Classifier-Degradation-Guidance上获得。
英文摘要：
Classifier-Free Guidance (CFG) is a cornerstone of modern text-to-image models, yet its reliance on a semantically vacuous null prompt ($\varnothing$) generates a guidance signal prone to geometric entanglement. This is a key factor limiting its precision, leading to well-documented failures in complex compositional tasks. We propose Condition-Degradation Guidance (CDG), a novel paradigm that replaces the null prompt with a strategically degraded condition, $\boldsymbol{c}_{\text{deg}}$. This reframes guidance from a coarse "good vs. null" contrast to a more refined "good vs. almost good" discrimination, thereby compelling the model to capture fine-grained semantic distinctions. We find that tokens in transformer text encoders split into two functional roles: content tokens encoding object semantics, and context-aggregating tokens capturing global context. By selectively degrading only the former, CDG constructs $\boldsymbol{c}_{\text{deg}}$ without external models or training. Validated across diverse architectures including Stable Diffusion 3, FLUX, and Qwen-Image, CDG markedly improves compositional accuracy and text-image alignment. As a lightweight, plug-and-play module, it achieves this with negligible computational overhead. Our work challenges the reliance on static, information-sparse negative samples and establishes a new principle for diffusion guidance: the construction of adaptive, semantically-aware negative samples is critical to achieving precise semantic control. Code is available at https://github.com/Ming-321/Classifier-Degradation-Guidance.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
