Metraipolis › archive › dc_d97695feeb50

Personalize at Test Time: Learning User Preferences for Image Generation — arXiv:2610.09015

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.09015
License (as archived)
CC0-1.0
Archived
2026-10-08T11:10:28.885Z
SHA-256 of the archived text
4f08fc77a2567f93980b9eb61e186a893a54bf05ca00dac5e8b642a43937f6d2
Stable URL of this page
https://metraipolis.com/documento/dc_d97695feeb50

How to cite

Personalize at Test Time: Learning User Preferences for Image Generation — arXiv:2610.09015 — arXiv (en), https://arxiv.org/abs/2610.09015 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d97695feeb50 (Metraipolis archive dc_d97695feeb50, sha256 4f08fc77a2567f93980b9eb61e186a893a54bf05ca00dac5e8b642a43937f6d2)
BibTeX
@misc{dcd97695feeb50,
  title = {Personalize at Test Time: Learning User Preferences for Image Generation — arXiv:2610.09015},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.09015},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d97695feeb50; sha256 4f08fc77a2567f93980b9eb61e186a893a54bf05ca00dac5e8b642a43937f6d2},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.09015. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Diffusion models can generate high-quality images, yet aligning their outputs with individual user preferences remains challenging. A key bottleneck is accurately modeling diverse user preferences from limited feedback. Existing approaches often rely on labor-intensive manual preference annotations or vision-language models (VLM) to extract preference information from user interaction histories, introducing substantial annotation or computational costs that limit scalability. We propose an approach that learns personalized reward models directly from users' historical image preference pairs. First, we use an autoencoder to compress hundreds of visual attributes into 50 attribute-anchored preference dimensions and train an evaluator to score images along these dimensions. We then represent each user's preferences as a linear combination of the shared dimension scores, estimating the user-specific weights by maximizing the likelihood of their observed pairwise preferences under the Bradley-Terry model. This formulation reduces per-user adaptation to optimizing a low-dimensional weight vector, simplifying optimization and enabling data-efficient personalization from sparse feedback. The learned personalized rewards guide image generation at inference time while keeping the diffusion model frozen. Experiments on real-user preference data show that our approach achieves approximately 77% held-out pairwise preference prediction accuracy and improves the alignment of generated images with individual user preferences.

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/2610.09015 — 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 2026-10-08T10:14:06.040Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).