Small-scale ViT-22B models outperform standard ViT under matched parameter counts on CIFAR, and a proposed ViTUnet runs image-to-image translation, though without strong quantitative validation.
Convit: Improving vision transformers with soft convolutional inductive biases
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
background 1representative citing papers
citing papers explorer
-
Deeper Inside Deep ViT
Small-scale ViT-22B models outperform standard ViT under matched parameter counts on CIFAR, and a proposed ViTUnet runs image-to-image translation, though without strong quantitative validation.