AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

ScalableViT: Rethinking the Context-Oriented Generalization of Vision Transformer

Rui Yang, Hailong Ma, Jie Wu, Yansong Tang, Xuefeng Xiao, Min Zheng, Xiu Li

ECCV 2022 · Architectures and efficiency › Attention and transformers · approach: Efficient and robust vision transformers · 70 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add SSA instead of WSA in all blocks (ablation)ScalableViT-S · ADE20K Semantic FPNmIoU: 38.9 -> 44.4Improved
add IWSA (LIM) instead of WSA in all blocks (ablation)ScalableViT-S · ADE20K Semantic FPNmIoU: 38.9 -> 43.5Improved
replace WSA with IWSA or SSA (ablation)ScalableViT-S · ImageNet-1Ktop-1: 82.4 -> 82.9Improved
add channel scalability (r_c>1) to SSA (ablation)ScalableViT-S · ADE20K Semantic FPNmIoU: 43.7 -> 44.4Improved
add channel scalability (r_c>1) to SSA (ablation)ScalableViT-S · ImageNet-1Ktop-1: 82.6 -> 82.9Improved
order IWSA before SSA rather than SSA before IWSA (ablation)ScalableViT-S · ImageNet-1Ktop-1: 83.0 -> 83.1Improved
use IWSA instead of SWSA alternated with SSA (ablation)ScalableViT-S · ImageNet-1Ktop-1: 82.9 -> 83.1Improved
add LIM to baseline (ablation)ScalableViT-S · ImageNet-1Ktop-1: 82.7 -> 83.0Improved
add LEM to baseline (ablation)ScalableViT-S · ImageNet-1Ktop-1: 82.7 -> 82.8Improved
add PEG to baseline (ablation)ScalableViT-S · ImageNet-1Ktop-1: 82.7 -> 82.9Improved
add LIM to baseline (ablation)ScalableViT-S · ADE20K Semantic FPNmIoU: 41.7 -> 43.7Improved
add PEG together with LIM to baseline (ablation)ScalableViT-S · ADE20K Semantic FPNmIoU: 41.7 -> 44.9Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy