AI Paper Atlas

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

Generating Representative Samples for Few-Shot Classification

Jingyi Xu, Hieu Le

CVPR 2022 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Alignment-based few-shot metric learning · 104 citations (OpenAlex)

Check 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 CVAE-generated semantic-conditioned featuresMeta-Baseline · miniImageNet 1-shotaccuracy (%): 63.17 ± 0.23 -> 69.96 ± 0.21Improved
Train the CVAE on representative samples instead of all base features (ablation)Meta-Baseline + SVAE · miniImageNet 1-shotaccuracy (%): 69.96 ± 0.21 -> 72.79 ± 0.19Improved
Add CVAE-generated semantic-conditioned featuresProtoNet · miniImageNet 1-shotaccuracy (%): 62.39 -> 73.01 ± 0.24Improved
Train the CVAE on representative samples instead of all base features (ablation)ProtoNet + SVAE · miniImageNet 1-shotaccuracy (%): 73.01 ± 0.24 -> 74.84 ± 0.23Improved
Add CVAE-generated semantic-conditioned featuresE3BM · miniImageNet 1-shotaccuracy (%): 64.09 ± 0.37 -> 73.07 ± 0.39Improved
Train the CVAE on representative samples instead of all base features (ablation)E3BM + SVAE · miniImageNet 1-shotaccuracy (%): 73.07 ± 0.39 -> 73.35 ± 0.37Improved
Add CVAE-generated semantic-conditioned featuresMeta-Baseline · tieredImageNet 1-shotaccuracy (%): 68.62 ± 0.27 -> 73.05 ± 0.24Improved
Train the CVAE on representative samples instead of all base features (ablation)Meta-Baseline + SVAE · tieredImageNet 1-shotaccuracy (%): 73.05 ± 0.24 -> 73.90 ± 0.24Improved
Add CVAE-generated semantic-conditioned featuresProtoNet · tieredImageNet 1-shotaccuracy (%): 68.23 -> 76.36 ± 0.65Improved
Train the CVAE on representative samples instead of all base features (ablation)ProtoNet + SVAE · tieredImageNet 1-shotaccuracy (%): 76.36 ± 0.65 -> 76.98 ± 0.65Improved
Add CVAE-generated semantic-conditioned featuresE3BM · tieredImageNet 1-shotaccuracy (%): 71.34 ± 0.41 -> 79.85 ± 0.43Improved
Train the CVAE on representative samples instead of all base features (ablation)E3BM + SVAE · tieredImageNet 1-shotaccuracy (%): 79.85 ± 0.43 -> 80.46 ± 0.43Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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