A-MoD routes tokens in Mixture-of-Depths vision transformers using attention maps from the prior layer, improving accuracy and convergence without extra trainable parameters.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Attention Is All You Need For Mixture-of-Depths Routing
A-MoD routes tokens in Mixture-of-Depths vision transformers using attention maps from the prior layer, improving accuracy and convergence without extra trainable parameters.