Metraipolis › archive › dc_a83086ab8025

Discriminative Score Function: Turning Pretrained Models into Functional Generative Priors — arXiv:2610.19743

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.19743
License (as archived)
CC0-1.0
Archived
2026-10-15T14:23:34.320Z
SHA-256 of the archived text
96794ec5568b7b41c31b7797abd60ac9e92a5b540627eab04bbaad9b0e876b20
Stable URL of this page
https://metraipolis.com/documento/dc_a83086ab8025

How to cite

Discriminative Score Function: Turning Pretrained Models into Functional Generative Priors — arXiv:2610.19743 — arXiv (en), https://arxiv.org/abs/2610.19743 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a83086ab8025 (Metraipolis archive dc_a83086ab8025, sha256 96794ec5568b7b41c31b7797abd60ac9e92a5b540627eab04bbaad9b0e876b20)
BibTeX
@misc{dca83086ab8025,
  title = {Discriminative Score Function: Turning Pretrained Models into Functional Generative Priors — arXiv:2610.19743},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.19743},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a83086ab8025; sha256 96794ec5568b7b41c31b7797abd60ac9e92a5b540627eab04bbaad9b0e876b20},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.19743. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Discriminative models are trained with task-specific objectives such as classification, detection, or representation learning, yet their practical utility depends on generalizing beyond the training task to unseen samples from the same domain. This suggests that a discriminative model may retain an implicit trace of the data manifold beyond its task labels. We ask whether this trace can be exposed as a generative signal without task-specific guidance. We introduce the Discriminative Score Function (DSF), which extracts an image-space generative field from the loss-gradient geometry of a pretrained discriminative model. DSF provides a training-free update rule that moves images toward model-consistent visual regions without retraining or task-specific targets. We apply DSF to diverse pretrained discriminative models, including ResNet classifiers, DETR detectors, and DINOv2 ViT encoders. DSF generates unconditional images from noise and supports conditional guidance, editing, inpainting, and explanation through the same field. DSF-generated samples also provide effective zero-shot quantization calibration when real images or labels are unavailable, preserving feature geometry better than data-free calibration baselines. These results show that discriminative training leaves a generative functional signal for synthesis and calibration. Project page: this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.19743 — 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 2026-10-15T13:53:35.899Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).