[Paper] Spectrally-Guided Diffusion Noise Schedules

Source: arXiv - 2603.19222v1

Overview

Denoising diffusion models are widely used for high-quality image and video generation. Their performance depends on noise schedules, which define the distribution of noise levels applied during training and the sequence of noise levels traversed during sampling. Noise schedules are typically handcrafted and require manual tuning across different resolutions. In this work, we propose a principled way to design per-instance noise schedules for pixel diffusion, based on the image’s spectral properties. By deriving theoretical bounds on the efficacy of minimum and maximum noise levels, we design “tight” noise schedules that eliminate redundant steps. During inference, we propose to conditionally sample such noise schedules. Experiments show that our noise schedules improve generative quality of single-stage pixel diffusion models, particularly in the low-step regime.

Key Contributions

This paper presents research in the following areas:

• cs.CV
• cs.LG

Methodology

Please refer to the full paper for detailed methodology.

Practical Implications

This research contributes to the advancement of cs.CV.

Authors

• Carlos Esteves
• Ameesh Makadia

Paper Information

• arXiv ID: 2603.19222v1
• Categories: cs.CV, cs.LG
• Published: March 19, 2026
• PDF: Download PDF