Metraipolis › archive › dc_7323aebca2fe

Less Supervision, Better Generalization: Weakly Supervised Fake Region Localization in Diffusion-Edited Images — arXiv:2609.36882

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.36882
License (as archived): CC0-1.0
Archived: 2026-10-07T18:38:37.233Z
SHA-256 of the archived text: 83d6f15803c789c5c587f94eddfab4d00d7c7554e04b9c522f27e89c6e074689
Stable URL of this page: https://metraipolis.com/documento/dc_7323aebca2fe

How to cite

Less Supervision, Better Generalization: Weakly Supervised Fake Region Localization in Diffusion-Edited Images — arXiv:2609.36882 — arXiv (en), https://arxiv.org/abs/2609.36882 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7323aebca2fe (Metraipolis archive dc_7323aebca2fe, sha256 83d6f15803c789c5c587f94eddfab4d00d7c7554e04b9c522f27e89c6e074689)

BibTeX

@misc{dc7323aebca2fe,
  title = {Less Supervision, Better Generalization: Weakly Supervised Fake Region Localization in Diffusion-Edited Images — arXiv:2609.36882},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.36882},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7323aebca2fe; sha256 83d6f15803c789c5c587f94eddfab4d00d7c7554e04b9c522f27e89c6e074689},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.36882. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Localizing AI-edited regions is essential for interpretable forensic analysis, but remains challenging due to subtle and spatially distributed artifacts that are misaligned with semantic or object boundaries. Existing approaches rely on pixel-level supervision from controlled editing pipelines, which is difficult to scale and can introduce misleading signals: artifacts frequently extend beyond annotated regions, while out-of-mask pixels are treated as authentic. This limits models' ability to capture transferable evidence and generalize across generators and datasets. To address these issues, we propose ReGFLoW, a Reconstruction-Guided Fake Localization framework under Weak supervision, which is the first weakly supervised approach for diffusion-edited fake region localization. ReGFLoW requires only real/fake labels at the image level and uses diffusion reconstruction errors as dense spatial guidance to inject them into both feature and score spaces. Furthermore, by artifact-centric multiple instance learning, ReGFLoW utilizes localized diffusion evidence without relying on semantic-affinity or boundary-based pseudo-mask priors. Extensive experiments show competitive cross-generator localization, while ReGFLoW outperforms all evaluated fully supervised baselines when evaluation includes both partially edited and fully synthetic images and in cross-dataset tests, without target-domain adaptation.

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/2609.36882 — 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 2026-10-07T18:04:24.771Z 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).