ProKeR treats CLIP cache models as Nadaraya-Watson estimators and fits a proximally regularized kernel ridge regression in an RKHS, reporting state-of-the-art training-free few-shot accuracy on 11 datasets.
Clip2scene: Towards label-efficient 3d scene understanding by clip
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
ProKeR: A Kernel Perspective on Few-Shot Adaptation of Large Vision-Language Models
ProKeR treats CLIP cache models as Nadaraya-Watson estimators and fits a proximally regularized kernel ridge regression in an RKHS, reporting state-of-the-art training-free few-shot accuracy on 11 datasets.