Metraipolis › archive › dc_70ea0a2a450b

Task Adaptation Through Guidance in Frozen Feature Upsampling — arXiv:2611.02854

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.02854
License (as archived)
CC0-1.0
Archived
2026-11-03T11:42:40.498Z
SHA-256 of the archived text
d295de1e650a6a05b9bc122d1a71b188f3d76389ff2d03e8662161a4a9a50665
Stable URL of this page
https://metraipolis.com/documento/dc_70ea0a2a450b

How to cite

Task Adaptation Through Guidance in Frozen Feature Upsampling — arXiv:2611.02854 — arXiv (en), https://arxiv.org/abs/2611.02854 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_70ea0a2a450b (Metraipolis archive dc_70ea0a2a450b, sha256 d295de1e650a6a05b9bc122d1a71b188f3d76389ff2d03e8662161a4a9a50665)
BibTeX
@misc{dc70ea0a2a450b,
  title = {Task Adaptation Through Guidance in Frozen Feature Upsampling — arXiv:2611.02854},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.02854},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_70ea0a2a450b; sha256 d295de1e650a6a05b9bc122d1a71b188f3d76389ff2d03e8662161a4a9a50665},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02854. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Adapting features before upsampling can improve dense prediction, but the gains may arise from changes in feature values, spatial aggregation, or both. We investigate this ambiguity by comparing matched residual adapters before and after a frozen JAFAR upsampler under a shared parent. We construct a readout-null intervention that preserves a frozen linear classifier's coarse logits while allowing spatial attention weights to adapt. Guidance/value exchanges on trained models and retrained fixed-guidance controls test the utility of this adaptation route. In three-seed experiments on PASCAL Visual Object Classes (VOC) and Cityscapes, adaptation before upsampling improves mean intersection over union (mIoU) by 0.807 and 1.413 percentage points, respectively, over adaptation after upsampling. Across both datasets and the JAFAR and AnyUp upsamplers, readout-null adaptation produces mean gains of 1.379-1.987 points over the corresponding frozen linear baselines. Retrained controls further support the benefit of adaptive spatial guidance. These findings explain how dense predictions can improve despite unchanged coarse classifier outputs and identify spatial guidance as a useful target when adapting frozen feature upsampling systems.

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/2611.02854 — 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 2026-11-03T11:06:37.597Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).