LEAP is an adaptive layer-skipping curriculum for ViT feature distillation that reports accuracy gains on ImageNet and retrieval tasks plus training compute savings.
FastDINOv2: Frequency based curriculum learning improves robustness and training speed.arXiv preprint arXiv:2507.03779, 2025
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
LEAP: Layer-skipping Efficiency via Adaptive Progression for Vision Transformer Distillation
LEAP is an adaptive layer-skipping curriculum for ViT feature distillation that reports accuracy gains on ImageNet and retrieval tasks plus training compute savings.