Predicting the outputs of several hidden layers of an EMA teacher, rather than only the final layer or pixels, substantially improves self-supervised ViT representations on ImageNet and downstream tasks.
MultiMAE : Multi-modal multi-task masked autoencoders
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Self-Distillation of Hidden Layers for Self-Supervised Representation Learning
Predicting the outputs of several hidden layers of an EMA teacher, rather than only the final layer or pixels, substantially improves self-supervised ViT representations on ImageNet and downstream tasks.