Everything at Every Scale: Scale-Invariant Diffusion with Continuous Super-Resolution
cs.CV, cond-mat.stat-mech, cs.AI, cs.LG
Submitted: 2026-05-25
Updated: 2026-11-05
Comments: 39 pages, 18 figures, NeurIPS 2026
Code: https://github.com/JazzyCH/SKILD
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Score-Based Generative Modeling through Stochastic Differential Equations
- On the Importance of Noise Scheduling for Diffusion Models
- Relay Diffusion: Unifying diffusion process across resolutions for image synthesis
- Renormalizing Diffusion Models
- Generative Modelling With Inverse Heat Dissipation
- Generative diffusion model with inverse renormalization group flows
- DCTdiff: Intriguing Properties of Image Generative Modeling in the DCT Space
- A Fourier Space Perspective on Diffusion Models
- Frequency Autoregressive Image Generation with Continuous Tokens
- Image Restoration with Mean-Reverting Stochastic Differential Equations
- I$^2$SB: Image-to-Image Schr\"odinger Bridge
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs