AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Self-Promoted Supervision for Few-Shot Transformer

Bowen Dong, Pan Zhou, Shuicheng Yan, Wangmeng Zuo

ECCV 2022 · Architectures and efficiency › Attention and transformers · approach: Hybrid and interpretable ViT designs

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 teacher ViT meta-training phase (f_g) (ablation)SUN (NesT) · miniImageNet 5-way 1-shotaccuracy: 49.23 -> 62.40Improved
add teacher ViT meta-training phase (f_g) (ablation)SUN (NesT) · miniImageNet 5-way 5-shotaccuracy: 66.57 -> 79.45Improved
add location-specific supervision (ablation)SUN (NesT) · miniImageNet 5-way 1-shotaccuracy: 62.40 -> 63.06Improved
add spatial-consistent augmentation (SCA) (ablation)SUN (NesT) · miniImageNet 5-way 1-shotaccuracy: 63.06 -> 64.50Improved
add background filtration (BGF) (ablation)SUN (NesT) · miniImageNet 5-way 1-shotaccuracy: 64.50 -> 64.84Improved
add meta-tuning phase (ablation)SUN (NesT) · miniImageNet 5-way 1-shotaccuracy: 64.84 -> 66.54Improved
add meta-tuning phase (ablation)SUN (NesT) · miniImageNet 5-way 5-shotaccuracy: 80.96 -> 82.09Improved
raise teacher drop path rate from 0.1 to 0.5 (ablation)teacher ViT · miniImageNet 5-way 1-shotaccuracy: 60.10 -> 62.40Improved
add CNN branch (inductive bias) to NesTMeta-Baseline (NesT) · miniImageNet 5-way 1-shotaccuracy: 54.57 -> 57.91Improved
add CNN distillation to NesTMeta-Baseline (NesT) · miniImageNet 5-way 1-shotaccuracy: 54.57 -> 55.79Improved
use local-attention ViT (Swin) instead of global attention (LeViT)Meta-Baseline · miniImageNet 5-way 1-shotaccuracy: 38.89 -> 54.63Improved
replace Meta-Baseline with SUN on ViTsLV-ViT · miniImageNet 5-way 1-shotaccuracy: 43.08 -> 59.00Improved

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