A contrastive pre-training method with separate losses for same-sample, same-class, and different-class relations achieves 1-shot state-of-the-art on miniImageNet, tieredImageNet, CIFAR-FS, and CUB.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Rethinking the Sample Relations for Few-Shot Classification
A contrastive pre-training method with separate losses for same-sample, same-class, and different-class relations achieves 1-shot state-of-the-art on miniImageNet, tieredImageNet, CIFAR-FS, and CUB.