AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Attribute Surrogates Learning and Spectral Tokens Pooling in Transformers for Few-Shot Learning

Yangji He, Weihan Liang, Dongyang Zhao, Hong-Yu Zhou, Weifeng Ge, Yizhou Yu, Wenqiang Zhang

CVPR 2022 · Architectures and efficiency › Attention and transformers · approach: Efficient and robust vision transformers · 64 citations (OpenAlex)

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 class and patch attribute-surrogate losses to DINO training (ablation)DINO ViT-S · miniImageNet5-way 1-shot accuracy: 61.57 ± 0.16 -> 71.27 ± 0.17Improved
Replace surrogate losses with cross-entropy supervision (ablation)DINO ViT-S · miniImageNet5-way 1-shot accuracy: 71.27 ± 0.17 -> 66.81 ± 0.17Hurt
Add the class-token surrogate loss (ablation)DINO with patch surrogate loss · miniImageNet5-way 1-shot accuracy: 63.17 ± 0.16 -> 71.27 ± 0.17Improved
Add the patch-token surrogate loss (ablation)DINO with class surrogate loss · miniImageNet5-way 1-shot accuracy: 68.95 ± 0.17 -> 71.27 ± 0.17Improved
Set patch-surrogate loss weight β to 1 instead of 0.1 (ablation)HCTransformers first student · miniImageNet5-way 1-shot accuracy: 71.27 ± 0.17 -> 61.45 ± 0.16Hurt
Set patch-surrogate loss weight β to 0.01 instead of 0.1 (ablation)HCTransformers first student · miniImageNet5-way 1-shot accuracy: 71.27 ± 0.17 -> 70.40 ± 0.16Hurt
Use 224²/8 rather than 56²/2 input/patch settings (ablation)HCTransformers first student · CIFAR-FS5-way 1-shot accuracy: 71.99 ± 0.17 -> 73.28 ± 0.16Improved
Use 224²/8 rather than 112²/4 input/patch settings (ablation)HCTransformers first student · CIFAR-FS5-way 1-shot accuracy: 72.80 ± 0.17 -> 73.28 ± 0.16Improved
Train the latter transformer sets one-by-one rather than end-to-end (ablation)HCTransformers · miniImageNetstage 2 1-shot accuracy: 74.74 ± 0.17 -> 74.40 ± 0.17Hurt
Train the latter transformer sets one-by-one rather than end-to-end (ablation)HCTransformers · miniImageNetstage 3 1-shot accuracy: 72.66 ± 0.18 -> 73.01 ± 0.18Improved
Supervise latter-stage patch tokens in addition to class tokens (ablation)HCTransformers later stages · miniImageNetstage 2 1-shot accuracy: 74.74 ± 0.17 -> 71.25 ± 0.17Hurt
Supervise class surrogate using global and local views instead of global views only (ablation)HCTransformers first student · miniImageNet5-way 1-shot accuracy: 71.27 ± 0.17 -> 69.37 ± 0.16Hurt

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