AI Paper Atlas

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

Benchmarking Spurious Bias in Few-Shot Image Classifiers

Guangtao Zheng, Wenqian Ye, Aidong Zhang

ECCV 2024 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Transductive and semantic-dictionary zero-shot · 1 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
construct test tasks with FewSTAB attribute-based selection instead of random samplingten few-shot classifiers · miniImageNet, tieredImageNet, CUB-200wAcc: RENet miniImageNet 5-shot 56.52 -> 36.19Hurt
add inter-class attribute-based sample selection to intra-class selection (ablation)FewSTAB · miniImageNet 5-way 5-shotaverage accuracy drop (%): 5.13 -> 13.30Improved
add likelihood-score based query sample selection (ablation)FewSTAB · miniImageNet 5-way 5-shotaverage accuracy drop (%): 13.30 -> 15.05Improved
switch from 1-shot to 5-shot test tasksFewSTAB · Spearman rank correlation wAcc-A vs wAcc-Rrank correlation: CUB-200 1.00 -> 0.94Mixed
increase training shots from 1-shot to 5-shot for meta-learning methodsmeta-learning FSC methods · FewSTAB tasks with 1/5/10 test shotswAcc-A gapMixed
replace ViT-GPT2 with BLIP as the attribute-detecting VLM (ablation)FewSTAB · miniImageNet 5-shotSpearman rank correlation of wAcc-A: 1.00No clear effect
increase test shots from 1 to 5 in FewSTABFewSTAB · miniImageNet, tieredImageNet, CUB-200accuracy gap (wAcc-R minus wAcc-A)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