Just Image Transformers with Late-Stage Pixel Refinement
cs.CV, cs.AI
Submitted: 2026-11-09
Updated: 2026-11-09
Code: https://github.com/deep-floyd/IF
Terminology
Sources
- Edify Image: High-Quality Image Generation with Pixel Space Laplacian Diffusion Models
- Asymmetric Flow Models
- PixelFlow: Pixel-Space Generative Models with Flow
- L2P: Unlocking Latent Potential for Pixel Generation
- Show the Signal, Hide the Noise: Spectral Forcing for Pixel-Space Diffusion
- HyperDiT: Hyper-Connected Transformers for High-Fidelity Pixel-Space Diffusion
- FrequencyBooster: Full-Frequency Modeling for High-Fidelity Pixel Diffusion
- PixelGen: Improving Pixel Diffusion with Perceptual Supervision
- Spectral Progressive Diffusion for Efficient Image and Video Generation
- Pixel-Space Post-Training of Latent Diffusion Models
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