arXiv · Text-to-Image
SEGA
Training-free resolution-extrapolation method for Diffusion Transformers (Spectral-Energy Guided Attention) — makes RoPE component scaling frequency-aware, adapting it to the spectral structure of the current latent. No retraining or architecture changes; improves structure/semantics/visual quality at higher resolutions on Flux and Qwen-Image.
Open the source ↗Similar Vision & Image releases
- Ideogram Vector
Ideogram · Vector Image Model · Week 44, 2026
- Fibo 2.0 Dev
Bria AI · Image Gen + Edit · Week 44, 2026
- AI Toolkit
ostris · Diffusion Fine-Tuning · Week 43, 2026
- MOSS-Image
OpenMOSS · Open Text-to-Image · Week 43, 2026
- FLUX 3 Image Klein
Black Forest Labs · Open Image Model · Week 43, 2026
- Iris-3B
Speridlabs · Pixel-Space Diffusion · Week 42, 2026
More from arXiv
- WavTTS
arXiv · Text-to-Speech · Week 23, 2026
- Stable-Layers
arXiv · Image Editing · Week 23, 2026
- Déjà View
arXiv · 3D Reconstruction · Week 23, 2026
- PIXLRelight
arXiv · Image Editing · Week 22, 2026
- SCOPE
arXiv · Video Understanding · Week 22, 2026
- InstructAV2AV
arXiv · Video Editing · Week 22, 2026
Published by Quanta AI Labs, an AI agency in India. Browse every week in AI News Weekly.