Metraipolis › archive › dc_bbd1c0431077
Rubric-CEPR: Self-Evolving Image Editing via Reward-Verified Self-Distillation — arXiv:2610.12469
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.12469
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-09T06:33:03.174Z
- SHA-256 of the archived text
a8b03545010801d16e035a29ae15ba7e517953c8f05b22a26e22962af5f939f5- Stable URL of this page
https://metraipolis.com/documento/dc_bbd1c0431077
How to cite
Rubric-CEPR: Self-Evolving Image Editing via Reward-Verified Self-Distillation — arXiv:2610.12469 — arXiv (en), https://arxiv.org/abs/2610.12469 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bbd1c0431077 (Metraipolis archive dc_bbd1c0431077, sha256 a8b03545010801d16e035a29ae15ba7e517953c8f05b22a26e22962af5f939f5)
BibTeX
@misc{dcbbd1c0431077,
title = {Rubric-CEPR: Self-Evolving Image Editing via Reward-Verified Self-Distillation — arXiv:2610.12469},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.12469},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bbd1c0431077; sha256 a8b03545010801d16e035a29ae15ba7e517953c8f05b22a26e22962af5f939f5},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.12469. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Instruction-guided image editors have become highly capable, yet improving them further still depends on human-edited training pairs or external reward models. Such supervision is costly to obtain and can reward plausible failures: a realistic output may leave the requested change undone or alter content that should be preserved. In this work, we strive to improve a pretrained image editor using only its own generations, without human-edited targets or an external training-time reward model. To this end, we propose a self-evolving framework, named Rubric-CEPR, that verifies the editor's own samples with its internal representations through a rubric-augmented Contrastive Edit-Preservation Reward (CEPR). A Planner proposes structured edit instructions from unlabeled images, the Editor samples multiple candidate edits, and a frozen Critic scores each candidate with decomposed rubric checks for edit realization, removal of the old state, and content preservation, using features already exposed by the editor. Non-compensatory gates reject infeasible candidates, and the best verified candidate is distilled into the editor through lightweight adapter training. On Qwen-Image-Edit, Rubric-CEPR improves ImgEdit from 4.36 to 4.60 (+5.5%), with a +24.9% gain on object isolation, and transfers to GEdit-Bench and Complex-Edit. The same procedure also improves Step1X-Edit by +7.8% on ImgEdit. We hope our approach will serve as a solid baseline for image editors that improve themselves from their own verified samples. Our code is publicly available at $\href{ this https URL }{\text{this URL}}$
Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.12469 — 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 2026-10-09T05:55:07.795Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).