Metraipolis › archive › dc_9729328aefca

Residual Calibration via Local Feature-Space Refinement — arXiv:2610.30652

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.30652
License (as archived): CC0-1.0
Archived: 2026-10-26T20:13:38.056Z
SHA-256 of the archived text: 0d4a6aacf2508ed8e2fc490e07bf3d97223f5a37b59f5183e99482d3a718fdb3
Stable URL of this page: https://metraipolis.com/documento/dc_9729328aefca

How to cite

Residual Calibration via Local Feature-Space Refinement — arXiv:2610.30652 — arXiv (en), https://arxiv.org/abs/2610.30652 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9729328aefca (Metraipolis archive dc_9729328aefca, sha256 0d4a6aacf2508ed8e2fc490e07bf3d97223f5a37b59f5183e99482d3a718fdb3)

BibTeX

@misc{dc9729328aefca,
  title = {Residual Calibration via Local Feature-Space Refinement — arXiv:2610.30652},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.30652},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9729328aefca; sha256 0d4a6aacf2508ed8e2fc490e07bf3d97223f5a37b59f5183e99482d3a718fdb3},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.30652. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Deep neural networks often struggle to provide reliable confidence estimates, making post-hoc calibration essential for reliable decision-making. Existing post-hoc methods largely rely on score-based transformations, which improve marginal calibration but often fail to capture spatially varying miscalibration in local regions of the learned feature space. To address this gap, we propose Local Adaptive Probability Estimation (LAPE), a post-hoc residual calibration framework. Specifically, LAPE treats the global calibrator output as a stable anchor and refines the individual prediction by leveraging non-parametric local evidence from neighboring calibration samples. As a plug-and-play refinement module, LAPE effectively utilizes the information about feature-space geometry and can reuse the calibration data that are used to construct the global calibrator. Furthermore, we extend LAPE to the setting with multi-view inputs through Cascade Feature Augmentation Fusion (CFAF). Extensive experiments demonstrate that the proposed methods can improve the calibration quality of various existing baselines and remain efficient and robust across different settings.

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); Machine Learning (stat.ML).
Pagina del lavoro: https://arxiv.org/abs/2610.30652 — 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 2026-10-26T19:49:03.446Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).