mlsanityData to 14 Oct 2026

Spectrally-Guided Diffusion Noise Schedules

arXiv 2603.19222 ↗First posted 19 Mar 2026cs.CV cs.LGcited by 5

On the map: Diffusion models → Diffusion foundations and sampling

Abstract (arXiv)

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 facts. Cited by 5 papers in the map (data through 14 Oct 2026). Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models. Builds on 52 of its 63 references that are in the map. First posted on arXiv 19 Mar 2026.

Builds on

52 of 63 references.

Closest ancestors · 29

Same direction, newest first

  1. 2025PixelDiT: Pixel Diffusion Transformers for Image Generation25 Nov
  2. 2025Diffusion Transformers with Representation Autoencoders13 Oct
  3. 2025PixNerd: Pixel Neural Field Diffusion31 Jul
  4. 2025A Fourier Space Perspective on Diffusion Models16 May
  5. 2025Improving the Diffusability of Autoencoders20 Feb
  6. 2025Shaping Inductive Bias in Diffusion Models through Frequency-Based Noise Control14 Feb
  7. 2025Spectral Analysis of Diffusion Models with Application to Schedule Design31 Jan
  8. 2024Improved Noise Schedule for Diffusion Training3 Jul
  9. 2024Multistep Distillation of Diffusion Models via Moment Matching6 Jun
  10. 2024Applying Guidance in a Limited Interval Improves Sample and Distribution Quality in Diffusion Models11 Apr
  11. 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesis5 Mar
  12. 2024Blue noise for diffusion models7 Feb
17 more
  1. 2023Diffusion Models With Learned Adaptive Noise20 Dec
  2. 2023One-step Diffusion with Distribution Matching Distillation30 Nov
  3. 2023SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis4 Jul
  4. 2023Common Diffusion Noise Schedules and Sample Steps are Flawed15 May
  5. 2023Consistency Models2 Mar
  6. 2023Understanding Diffusion Objectives as the ELBO with Simple Data Augmentation1 Mar
  7. 2023Simple diffusion: End-to-end diffusion for high resolution images26 Jan
  8. 2023On the Importance of Noise Scheduling for Diffusion Models26 Jan
  9. 2022On Distillation of Guided Diffusion Models6 Oct
  10. 2022Classifier-Free Diffusion Guidance26 Jul
  11. 2022Elucidating the Design Space of Diffusion-Based Generative Models1 Jun
  12. 2021High-Resolution Image Synthesis with Latent Diffusion Models20 Dec
  13. 2021Variational Diffusion Models1 Jul
  14. 2021Cascaded Diffusion Models for High Fidelity Image Generation30 May
  15. 2021Diffusion Models Beat GANs on Image Synthesis11 May
  16. 2021Improved Denoising Diffusion Probabilistic Models18 Feb
  17. 2020Denoising Diffusion Probabilistic Models19 Jun

Other references · 23

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  4. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  5. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,058
  6. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,585
  7. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,355
  8. 2022Flow Matching for Generative Modelingcited by 5,702
  9. 2022Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flowcited by 3,700
  10. 2016Improved Techniques for Training GANscited by 2,970
  11. 2017FiLM: Visual Reasoning with a General Conditioning Layercited by 2,333
  12. 2023Align your Latents: High-Resolution Video Synthesis with Latent Diffusion Modelscited by 1,476
11 more
  1. 2022Building Normalizing Flows with Stochastic Interpolantscited by 995
  2. 2019Improved Precision and Recall Metric for Assessing Generative Modelscited by 944
  3. 2023Language Model Beats Diffusion -- Tokenizer is Key to Visual Generationcited by 640
  4. 2025Mean Flows for One-step Generative Modelingcited by 561
  5. 2021Generating Images with Sparse Representationscited by 266
  6. 2022Scalable Adaptive Computation for Iterative Generationcited by 195
  7. 2022Generative Modelling With Inverse Heat Dissipationcited by 167
  8. 2023SwiftBrush: One-Step Text-to-Image Diffusion Model with Variational Score Distillationcited by 110
  9. 2024Improving the Training of Rectified Flowscited by 108
  10. 2025PixelFlow: Pixel-Space Generative Models with Flowcited by 97
  11. 2025Learnings from Scaling Visual Tokenizers for Reconstruction and Generationcited by 32

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 5.

5 papers.

  1. 2026SANTS: A State-Adaptive Scheduler for World Action Modelscited by 7
  2. 2026SALD: Self-Referenced Advantage Learning for Diffusion Modelscited by 1
  3. 2026When does training on downscaled images yield the same gradients?cited by 0
  4. 2026PRISM: Principled Reference Identification for Schrodinger Bridge Modelcited by 0
  5. 2026Beyond the Manifold Hypothesis: Hybrid Spectral Parameterizations for Flow Matchingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Spectral Progressive Diffusion for Efficient Image and Video Generation
  2. 2025A Comprehensive Review on Noise Control of Diffusion Model
  3. 2026Show the Signal, Hide the Noise: Spectral Forcing for Pixel-Space Diffusion
  4. 2026Energy-Conditioned Noise Schedule and Whitening for Spectral Diffusion
  5. 2024Enhancing Sample Generation of Diffusion Models using Noise Level Correction
  6. 2024Edge-preserving noise for diffusion models
  7. 2024Constant Rate Scheduling: A General Framework for Optimizing Diffusion Noise Schedule via Distributional Change
  8. 2024Training-Free Adaptive Diffusion with Bounded Difference Approximation Strategy