# NoiseShift: Resolution-Aware Noise Recalibration for Better Low-Resolution Image Generation arXiv: https://arxiv.org/abs/2510.02307 ยท HTML: https://mlsanity.com/paper/2510.02307/ ## Key facts - No citing papers in the map yet (data through 14 Oct 2026). - Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models. - Builds on 32 of its 41 references that are in the map. - First posted on arXiv 2 Oct 2025. ## Placed in - Diffusion models / [Diffusion foundations and sampling](https://mlsanity.com/diffusion/diffusion-foundations-and-sampling/) ## Builds on (most cited first, 30 of 32) - [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015) - [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021) - [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022) - [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022) - [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023) - [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023) - [BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation](https://mlsanity.com/paper/2201.12086/) (2022) - [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024) - [LAION-5B: An open large-scale dataset for training next generation image-text models](https://mlsanity.com/paper/2210.08402/) (2022) - [Deep Learning Face Attributes in the Wild](https://mlsanity.com/paper/1411.7766/) (2014) - [CLIPScore: A Reference-free Evaluation Metric for Image Captioning](https://mlsanity.com/paper/2104.08718/) (2021) - [DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps](https://mlsanity.com/paper/2206.00927/) (2022) - [DPM-Solver++: Fast Solver for Guided Sampling of Diffusion Probabilistic Models](https://mlsanity.com/paper/2211.01095/) (2022) - [MultiDiffusion: Fusing Diffusion Paths for Controlled Image Generation](https://mlsanity.com/paper/2302.08113/) (2023) - [Simple diffusion: End-to-end diffusion for high resolution images](https://mlsanity.com/paper/2301.11093/) (2023) - [OmniGen: Unified Image Generation](https://mlsanity.com/paper/2409.11340/) (2024) - [On the Importance of Noise Scheduling for Diffusion Models](https://mlsanity.com/paper/2301.10972/) (2023) - [ScaleCrafter: Tuning-free Higher-Resolution Visual Generation with Diffusion Models](https://mlsanity.com/paper/2310.07702/) (2023) - [FoleyCrafter: Bring Silent Videos to Life with Lifelike and Synchronized Sounds](https://mlsanity.com/paper/2407.01494/) (2024) - [DemoFusion: Democratising High-Resolution Image Generation With No ](https://mlsanity.com/paper/2311.16973/) (2023) - [Alleviating Exposure Bias in Diffusion Models through Sampling with Shifted Time Steps](https://mlsanity.com/paper/2305.15583/) (2023) - [FouriScale: A Frequency Perspective on Training-Free High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.12963/) (2024) - [Training-free Diffusion Model Adaptation for Variable-Sized Text-to-Image Synthesis](https://mlsanity.com/paper/2306.08645/) (2023) - [Make a Cheap Scaling: A Self-Cascade Diffusion Model for Higher-Resolution Adaptation](https://mlsanity.com/paper/2402.10491/) (2024) - [ElasticDiffusion: Training-free Arbitrary Size Image Generation through Global-Local Content Separation](https://mlsanity.com/paper/2311.18822/) (2023) - [Any-Size-Diffusion: Toward Efficient Text-Driven Synthesis for Any-Size HD Images](https://mlsanity.com/paper/2308.16582/) (2023) - [FreeScale: Unleashing the Resolution of Diffusion Models via Tuning-Free Scale Fusion](https://mlsanity.com/paper/2412.09626/) (2024) - [DiffCLIP: Leveraging Stable Diffusion for Language Grounded 3D Classification](https://mlsanity.com/paper/2305.15957/) (2023) - [ResAdapter: Domain Consistent Resolution Adapter for Diffusion Models](https://mlsanity.com/paper/2403.02084/) (2024) - [Instant 3D Human Avatar Generation using Image Diffusion Models](https://mlsanity.com/paper/2406.07516/) (2024) ## Abstract (arXiv) Text-to-image diffusion models often degrade when sampled at resolutions outside the final training resolution set. Prior work has largely emphasized higher resolution generation, enabling pretrained diffusion models to extrapolate beyond the resolutions seen during training. In this work, we instead target lower-resolution generation, performing inference at reduced resolution to significantly cut computational cost. We show that network conditioning of the noise level induces a train-test mismatch that directly degrades low-resolution generation: the same scheduled noise level can correspond to a different perceptual corruption level at lower resolutions, mis-calibrating the denoiser timestep and noise embedding. To this end, we propose NoiseShift, a training-free recalibration method that keeps the original noise sampling schedule unchanged and instead re-indexes the noise conditioning of the denoiser to restore local forward-reverse consistency. Using a lightweight coarse-to-fine calibration on a small set of image-text pairs, NoiseShift learns a resolution-specific mapping from scheduler noise to conditioning noise, reducing train-test mismatch and improving lower-resolution generation quality. When NoiseShift is applied to Stable Diffusion 3 (SD3), Stable Diffusion 3.5 (SD3.5), and Flux-Dev, generation quality at low resolutions improves consistently. Particularly, SD3 generation at 128x128 resolution gets an improved FID score from 203 to 171, and SD3.5 gets an improved FID score from 310 to 277 on LAION-COCO. Even Flux-Dev which already implements a complementary time-shifting strategy gets a modest boost from NoiseShift with an improved FID score from 120 to 113 at 64x64 resolution. More importantly, NoiseShift achieves such improvements with minimal implementation changes and no additional inference overhead.