Research
Can We Change the Stroke Size for Easier Diffusion?
arXiv:2603.26783v2 Announce Type: replace Abstract: Diffusion models can be challenged in the low signal-to-noise regime, where they have to make pixel-level predictions despite the presence of high n
arXiv:2603.26783v2 Announce Type: replace Abstract: Diffusion models can be challenged in the low signal-to-noise regime, where they have to make pixel-level predictions despite the presence of high noise. The geometric intuition is akin to using the finest stroke for oil painting throughout, which may be ineffective. We therefore study stroke-size control as a controlled intervention that changes the effective roughness of the supervised target, predictions and perturbations across timesteps, in an attempt to ease the low signal-to-noise challenge.
Related
- PixelDiT: Pixel Diffusion Transformers for Image Generation
- Edge-preserving noise for diffusion models
- Iterative Inference-time Scaling with Adaptive Frequency Steering for Image Super-Resolution
- RectifiedHR: Enable Efficient High-Resolution Synthesis via Energy Rectification
- Latent Wavelet Diffusion For Ultra-High-Resolution Image Synthesis
Source: arXiv cs.CV | 2026-04-29