AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

From Channel Bias to Feature Redundancy: Uncovering the “Less is More” Principle in Few-Shot Learning

Jie Zhang, Xu Luo, Lianli Gao, Difan Zou, Heng Tao Shen, Jingkuan Song

TPAMI 2026 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Embedding adaptation for few-shot classification · 17 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
apply simple channel-wise transformation phi_k with k=1.3 at test timefew-shot classifiers across backbones · Meta-Dataset/BSCD-FSL/DomainNet 5-way 5-shotaccuracy: 0.5-7.5% on averageImproved
apply simple transformation when train and test distributions are very similarImageNet-trained models · miniImageNet and MSCOCO testaccuracyHurt
replace simple transformation with oracle MMC (ablation)PN NCC · binary 5-shot tasks (avg of 10 datasets)accuracy: 83.1 -> 87.7Improved
replace simple transformation with oracle MMC (ablation)S2M2 linear classifier · binary 5-shot tasks (avg of 10 datasets)accuracy: 87.8 -> 91.7Improved
add simple transformation to unadapted features (ablation)PN NCC · binary 5-shot tasks (avg of 10 datasets)accuracy: 80.7 -> 83.1Improved
apply simple transformation to a fine-tuned feature extractor (ablation)fine-tuning on miniImageNet shots · miniImageNet shot analysisaccuracyNo clear effect
increase number of shots for linear classifier with transformation (ablation)linear classifier · miniImageNet shot analysis (up to 400 shots)accuracy gainMixed
mask unimportant feature dimensions, keeping about 110 (linear probing) or 255 (finetuning) (ablation)ImageNet ResNet-50 features · Aircraft 5-way 1-shottest error: around 110 for linear probing and 255 for finetuningImproved
add redundant second dimension to features (ablation)1-shot LP/NCC · two-dimensional Gaussian toyaccuracy: 90.83 -> 75.87Hurt
add second dimension with 500 shots (ablation)linear probing · two-dimensional Gaussian toyaccuracy: 95.20 -> 97.36Improved
add second dimension with 500 shots (ablation)NCC · two-dimensional Gaussian toyaccuracy: 95.21 -> 84.37Hurt
estimate feature importance from 5 randomly cropped augmented views per support image (ablation)feature importance estimation · few-shot tasksestimation variance and accuracyImproved

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