Metraipolis › archive › dc_0d0bace7abb5
Feed the Panel Dimensions, Not Verdicts: Rubric-Decomposed Fusion of Vision-Language Aesthetic Judges — arXiv:2609.27110
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.27110
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T10:05:34.059Z
- SHA-256 of the archived text
c3cbb0c3b3a30af5516571e6845f51f6b2be19abbf52a60fe3c992f889a0642e- Stable URL of this page
https://metraipolis.com/documento/dc_0d0bace7abb5
How to cite
Feed the Panel Dimensions, Not Verdicts: Rubric-Decomposed Fusion of Vision-Language Aesthetic Judges — arXiv:2609.27110 — arXiv (en), https://arxiv.org/abs/2609.27110 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0d0bace7abb5 (Metraipolis archive dc_0d0bace7abb5, sha256 c3cbb0c3b3a30af5516571e6845f51f6b2be19abbf52a60fe3c992f889a0642e)
BibTeX
@misc{dc0d0bace7abb5,
title = {Feed the Panel Dimensions, Not Verdicts: Rubric-Decomposed Fusion of Vision-Language Aesthetic Judges — arXiv:2609.27110},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.27110},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0d0bace7abb5; sha256 c3cbb0c3b3a30af5516571e6845f51f6b2be19abbf52a60fe3c992f889a0642e},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.27110. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Vision-language models (VLMs) are deployed as zero-shot judges of image aesthetics, and panels of several models are recommended, on thin evidence, as the way to make such judges reliable. On two human-rated datasets, EVA and PARA, we find that a panel of holistic judges never significantly beats its best member, whether the verdicts are averaged or fused by a learned combiner. What a panel is worth depends on what it is fed. We therefore have each model score each image on the five dimensions of a frozen, human-written rubric and fuse those scores, alongside each model's verdict, across model families with an out-of-fold combiner. The dimension scores measure what their labels claim: with the overall human score partialled out, a dimension prompt carries more attribute-specific information than the holistic prompt in 28 of 30 model-attribute cells. Fused, they beat the best single VLM in all ten three-family panels on EVA (against that best single model, +0.07 Spearman rho for the strongest trio and +0.10 for the pre-declared one, and +0.06 and +0.07 when averaged over twenty fold partitions; against the panel mean, the primary test gives +0.118 on its EVA design set), and on PARA they reach parity under Spearman rho and a small, non-significant loss under Kendall tau-b, where one model already captures 85% of the human noise ceiling. It is not a feature-count artefact: giving the same combiner an equal number of pure holistic columns, split from the same repetitions, does not reproduce it. The gain costs a few hundred labels, which do not transfer between datasets, and 4.8x the API calls on EVA; we report it with paired bootstraps and Kendall tau-b, alongside a failed pre-registration and the configurations that lost. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-24 in cs.CL). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Computation and Language (cs.CL); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.27110 — 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 2026-10-08T09:11:49.677Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-24?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).