English

SEGA: Spectral-Energy Guided Attention for Resolution Extrapolation in Diffusion Transformers

Computer Vision and Pattern Recognition 2026-05-22 v1

Abstract

Diffusion transformers (DiTs) have emerged as a dominant architecture for text-to-image generation, yet their performance drops when generating at resolutions beyond their training range. Existing training-free approaches mitigate this by modifying inference-time attention behavior, often through Rotary Position Embeddings (RoPE) extrapolation combined with attention scaling. However, these strategies apply a uniform and content-agnostic scaling across RoPE components with distinct frequency characteristics, inducing a trade-off between preserving global structure and recovering fine detail. We introduce SEGA, a training-free method that dynamically scales attention across RoPE components according to the latent's spatial-frequency structure at each denoising step. This adaptive scaling improves both structural coherence and fine-detail fidelity. Experiments show that SEGA consistently improves high-resolution synthesis across multiple target resolutions, outperforming state-of-the-art training-free baselines.

Keywords

Cite

@article{arxiv.2605.22668,
  title  = {SEGA: Spectral-Energy Guided Attention for Resolution Extrapolation in Diffusion Transformers},
  author = {Javad Rajabi and Kimia Shaban and Koorosh Roohi and David B. Lindell and Babak Taati},
  journal= {arXiv preprint arXiv:2605.22668},
  year   = {2026}
}

Comments

27 pages, 14 figures. Project page: https://rajabi2001.github.io/sega/

R2 v1 2026-07-22T07:26:38.056Z