Metraipolis › archive › dc_1bd2ee3a014a

From Evaluation to Evolution: Benchmarking Image Editing and Evolving Image Models — arXiv:2610.41210

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.41210
License (as archived)
CC0-1.0
Archived
2026-11-03T07:33:14.171Z
SHA-256 of the archived text
3e806a95472d330afe110ec5bd4b0619ac39fd7b23fb1d117b8152c4d3899907
Stable URL of this page
https://metraipolis.com/documento/dc_1bd2ee3a014a

How to cite

From Evaluation to Evolution: Benchmarking Image Editing and Evolving Image Models — arXiv:2610.41210 — arXiv (en), https://arxiv.org/abs/2610.41210 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_1bd2ee3a014a (Metraipolis archive dc_1bd2ee3a014a, sha256 3e806a95472d330afe110ec5bd4b0619ac39fd7b23fb1d117b8152c4d3899907)
BibTeX
@misc{dc1bd2ee3a014a,
  title = {From Evaluation to Evolution: Benchmarking Image Editing and Evolving Image Models — arXiv:2610.41210},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.41210},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_1bd2ee3a014a; sha256 3e806a95472d330afe110ec5bd4b0619ac39fd7b23fb1d117b8152c4d3899907},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.41210. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Real-world image editing requires coordinating multiple visual transformations with content preservation and reference constraints, yet aggregate scores can readily obscure unmet local requirements. Effective evaluation must therefore identify local failures and translate them into priorities for subsequent model improvement. We introduce WeGenEditBench, comprising 2,740 evaluation instances and 148 task types across single-image, text-intensive, and multi-reference editing. Pre-generated checklists shared across models decompose each request into atomic requirements; item-level deductions and visual evidence are established first and then summarized into scores for edit execution, preservation, visual quality, and task-specific correctness. To support the experience-acquisition component of recursive self-improvement (RSI), we develop WeGenEvoStudio, a unified multi-agent harness for RSI-oriented learning that integrates model-conditioned failure diagnosis, retrieval-grounded data construction, supervised adaptation, and re-evaluation. Among 15 evaluated configurations, GPT-Image-2 leads single-image editing with an Overall score of 4.076, while Seedream-5.0-Pro leads text-intensive editing at 4.291. Evolution experiments across four generation and editing benchmarks demonstrate substantial targeted gains: on their report-selected Top-5 single-image targets, FLUX.2 [dev] improves Overall from 1.083 to 3.517, while Qwen-Image-Edit-2511 gains 13.2%. Together, our benchmark and harness provide a foundation for developing more reliable image models through interpretable evaluation and feedback-driven learning.

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).
Pagina del lavoro: https://arxiv.org/abs/2610.41210 — 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 2026-11-03T07:29:33.714Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).