AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

EnViT: Enhancing the Performance of Early-Exit Vision Transformers via Exit-Aware Structured Dropout-Enabled Self-Distillation

Yonghao Dong, Qiang He, Penghong Rui, Zhenzhe Zheng, Zhao Li, Feifei Chen, Hai Jin, Yun Yang

AAAI 2026 · Architectures and efficiency › Attention and transformers · approach: Hybrid and interpretable ViT designs · 0 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 exit-aware structured dropout (ablation)early-exit ViT-B/16 baseline · CIFAR-100 exit s2accuracy: 75.61 -> 76.23Improved
add exit-aware structured dropout (ablation)early-exit ViT-B/16 baseline · CIFAR-100 final exit s6accuracy: 89.40 -> 89.88Improved
add self-distillation without structured dropout (ablation)early-exit ViT-B/16 baseline · CIFAR-100 exit s2accuracy: 75.61 -> 78.05Improved
add self-distillation without structured dropout (ablation)early-exit ViT-B/16 baseline · CIFAR-100 final exit s6accuracy: 89.40 -> 87.73Hurt
add self-distillation without structured dropout (ablation)early-exit ViT-B/16 baseline · CIFAR-100 exit s5accuracy: 88.85 -> 87.68Hurt
add self-distillation without structured dropout (ablation)early-exit ViT-B/16 baseline · CIFAR-100 exit s4accuracy: 87.21 -> 86.51Hurt
combine structured dropout and self-distillation (full EnViT) (ablation)early-exit ViT-B/16 baseline · CIFAR-100 exit s1accuracy: 60.48 -> 61.94Improved
combine structured dropout and self-distillation (full EnViT) (ablation)early-exit ViT-B/16 baseline · CIFAR-100 exit s2accuracy: 75.61 -> 79.02Improved
add structured dropout on top of self-distillation (ablation)EnViT-KD · CIFAR-100 final exit s6accuracy: 87.73 -> 89.94Improved
add self-distillation on top of structured dropout (ablation)EnViT-SD · CIFAR-100 exit s2accuracy: 76.23 -> 79.02Improved
replace LGViT with EnViTLGViT · CIFAR-100top-1 accuracy: 88.14% -> 89.23%Improved
replace ViT-EE with EnViTViT-EE · ImageNet-1ktop-1 accuracy: 77.57% -> 80.45%Improved

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