Metraipolis › archive › dc_ab38070fc294

Generalizable AI-Generated Image Detection with Embedding-Space Explanations — arXiv:2611.04111

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.04111
License (as archived)
CC0-1.0
Archived
2026-11-04T18:36:27.368Z
SHA-256 of the archived text
2fae92c44bcdd180eeaa76b1eed6545ca1555a2062dd16e3b030ecdd456ed9f1
Stable URL of this page
https://metraipolis.com/documento/dc_ab38070fc294

How to cite

Generalizable AI-Generated Image Detection with Embedding-Space Explanations — arXiv:2611.04111 — arXiv (en), https://arxiv.org/abs/2611.04111 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ab38070fc294 (Metraipolis archive dc_ab38070fc294, sha256 2fae92c44bcdd180eeaa76b1eed6545ca1555a2062dd16e3b030ecdd456ed9f1)
BibTeX
@misc{dcab38070fc294,
  title = {Generalizable AI-Generated Image Detection with Embedding-Space Explanations — arXiv:2611.04111},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.04111},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ab38070fc294; sha256 2fae92c44bcdd180eeaa76b1eed6545ca1555a2062dd16e3b030ecdd456ed9f1},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.04111. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

AI-generated images have reached unprecedented realism, while instruction-driven AI editors re-synthesize images for local changes, leaving no splice boundary to localize. Most detectors are black boxes with unverifiable verdicts and generalize poorly because adapting a backbone on the detection distribution memorizes generator-specific fingerprints that quickly become obsolete. Specialist detectors built on pure-vision representations, particularly DINOv3, generalize well across generators, but explainable methods forgo this advantage: jointly training one multimodal model for classification and language generation subjects its visual representation to the language objective. We therefore require that explanation training cannot alter the verdict. We introduce FOX, which first builds and locks a specialist DINOv3 detector, then uses its representation to guide a lightweight VLM toward forensic explanation. A joint-embedding predictor maps locked visual evidence into a frozen text encoder's semantic space, producing guidance that the VLM verbalizes on demand. FOX achieves 97.4 mean AUC across 14 unseen generators, 69.8 BERTScore on XPlainVerse, and 56.1 on LOKI.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI); Multimedia (cs.MM).
Pagina del lavoro: https://arxiv.org/abs/2611.04111 — 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 2026-11-04T18:32:37.415Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).