Metraipolis › archive › dc_01ec61eb40dd

ChatGPT Images 2.5 on Forgery Tasks: Testing Advertised Improvements Against Known Answers — arXiv:2609.13617

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.13617
License (as archived)
CC0-1.0
Archived
2026-10-16T01:52:41.230Z
SHA-256 of the archived text
afe504f28beb1e18614f9c3aadfaf45f435912863818ffa8fa7d6794e8da3ba0
Stable URL of this page
https://metraipolis.com/documento/dc_01ec61eb40dd

How to cite

ChatGPT Images 2.5 on Forgery Tasks: Testing Advertised Improvements Against Known Answers — arXiv:2609.13617 — arXiv (en), https://arxiv.org/abs/2609.13617 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_01ec61eb40dd (Metraipolis archive dc_01ec61eb40dd, sha256 afe504f28beb1e18614f9c3aadfaf45f435912863818ffa8fa7d6794e8da3ba0)
BibTeX
@misc{dc01ec61eb40dd,
  title = {ChatGPT Images 2.5 on Forgery Tasks: Testing Advertised Improvements Against Known Answers — arXiv:2609.13617},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.13617},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_01ec61eb40dd; sha256 afe504f28beb1e18614f9c3aadfaf45f435912863818ffa8fa7d6794e8da3ba0},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.13617. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

OpenAI released ChatGPT Images 2.5 on 8 September 2026, advertising more precise local edits, better consistency across edits, more faithful reference products and sharper detail. We evaluate these claims on four forgery tasks with answers fixed in advance: receipt-field alteration, repeated editing, product placement and small-print rendering. GPT-Image-2 provides same-week baselines at a cheaper and a more expensive tier. A limited improvement appears in receipt editing. After alignment, OCR detects changes to surrounding text in 31.7% of Flare outputs, against 44.2% for the cheaper baseline. This gain is concentrated on CORD receipts and sensitive to shifts of a pixel or less; the forged value itself is no more often correct. Repeated editing and fine print show no measurable gain. Product codes become more legible mainly because Images 2.5 draws the product larger. Defence outcomes change little: localisation remains weak for both generations. A detector that flags 68.6% of controlled benchmark images flags only 35.9% of images posted online. Advertised improvements therefore transfer unevenly to the tested forgery capabilities, while substantial detection limitations remain.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-15 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.13617 — 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 2026-10-16T01:25:23.511Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).