Metraipolis › archive › dc_39aabe482d23
Attention-Scoped Guidance: Training-Free Spatial Control for Image Editing — arXiv:2609.37492
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.37492
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T18:41:19.607Z
- SHA-256 of the archived text
b8c489ef76afac46af9fe450be51d6cb1e2f1e7eb3733db151b08ccdee9ce878- Stable URL of this page
https://metraipolis.com/documento/dc_39aabe482d23
How to cite
Attention-Scoped Guidance: Training-Free Spatial Control for Image Editing — arXiv:2609.37492 — arXiv (en), https://arxiv.org/abs/2609.37492 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_39aabe482d23 (Metraipolis archive dc_39aabe482d23, sha256 b8c489ef76afac46af9fe450be51d6cb1e2f1e7eb3733db151b08ccdee9ce878)
BibTeX
@misc{dc39aabe482d23,
title = {Attention-Scoped Guidance: Training-Free Spatial Control for Image Editing — arXiv:2609.37492},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.37492},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_39aabe482d23; sha256 b8c489ef76afac46af9fe450be51d6cb1e2f1e7eb3733db151b08ccdee9ce878},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.37492. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Instruction-guided image editing should change what the instruction names and leave the rest of the image untouched. In dual classifier-free guidance (CFG), an editor combines two directions at every denoising step, one that pushes toward the instructed edit and one that pulls back toward the source image, using global weights. We introduce Attention-Scoped Guidance (ASG), a sampler wrapper that makes these weights spatial. It reads a soft support map from the instruction attention that the editor already computes, then weakens text guidance where support is low and strengthens image anchoring where support is high. The wrapper requires no training, no external mask, and no additional network evaluation. On the full MagicBrush and PIE-Bench++ splits, ASG improves preservation-oriented metrics, leading three of four MagicBrush metrics and PIE-Bench++ background PSNR. A dose-matched control that removes the spatial placement loses up to 0.73 CLIP on PIE-Bench++, confirming that the spatial allocation itself carries the gain. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-30 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.37492 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-07T18:04:24.877Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).