Research

Paper

AI LLM March 11, 2026

Guiding Diffusion Models with Semantically Degraded Conditions

Authors

Shilong Han, Yuming Zhang, Hongxia Wang

Abstract

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.

Metadata

arXiv ID: 2603.10780
Provider: ARXIV
Primary Category: cs.CV
Published: 2026-03-11
Fetched: 2026-03-12 04:21

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.10780v1</id>\n    <title>Guiding Diffusion Models with Semantically Degraded Conditions</title>\n    <updated>2026-03-11T13:54:35Z</updated>\n    <link href='https://arxiv.org/abs/2603.10780v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.10780v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>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.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n    <published>2026-03-11T13:54:35Z</published>\n    <arxiv:comment>Accepted to CVPR 2026</arxiv:comment>\n    <arxiv:primary_category term='cs.CV'/>\n    <author>\n      <name>Shilong Han</name>\n    </author>\n    <author>\n      <name>Yuming Zhang</name>\n    </author>\n    <author>\n      <name>Hongxia Wang</name>\n    </author>\n  </entry>"
}