Guidance in the Frequency Domain Enables High-Fidelity Sampling at Low CFG Scales

ICLR 2026 Conference Submission12938 Authors

Keywords: diffusion models, classifier-free guidance

TL;DR: We show that applying classifier-free guidance in the frequency domain substantially improves the quality at low guidance scales, while inherently avoiding the shortcomings associated with high guidance values.

Abstract: Classifier-free guidance (CFG) has become an essential component of modern conditional diffusion models. Although highly effective in practice, the underlying mechanisms by which CFG enhances quality, detail, and prompt alignment are not fully understood. This paper presents a novel perspective on CFG by analyzing its effects in the frequency domain, showing that low and high frequencies have distinct impacts on generation quality. Specifically, low-frequency guidance governs global structure and condition alignment, while high-frequency guidance mainly enhances visual fidelity. However, applying a uniform scale across all frequencies---as is done in standard CFG---leads to oversaturation and reduced diversity at high scales and degraded visual quality at low scales. Based on these insights, we propose frequency-decoupled guidance (FDG), an effective approach that decomposes CFG into low- and high-frequency components and applies separate guidance strengths to each component. FDG improves image quality at low guidance scales and avoids the drawbacks of high CFG scales by design, i.e., retaining the benefits of CFG over unguided generation without its common failures. Through extensive experiments across multiple datasets and models, we demonstrate that FDG consistently enhances sample fidelity while preserving diversity, leading to improved FID and recall compared to CFG, establishing our method as a plug-and-play alternative to standard classifier-free guidance.

Primary Area: generative models

Submission Number: 12938