Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Parameterless Transductive Feature Re-representation for Few-Shot Learning
Wentao Cui, Yuhong Guo
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 changed | Where | Measured | Effect |
|---|---|---|---|
| apply parameterless re-representation (RR) at inference on top of inductive/meta-learning baselines | ProtoNet, Matching, Relation (ResNet-18) · mini-ImageNet | accuracy: +2% to +4% (1-shot); +1.3% to +1.9% (5-shot) | Improved |
| add SSL loss during meta-training (MT vs RR) | ProtoNet, Matching, Relation · mini-ImageNet | accuracy: slight but consistent gains over RR version | Improved |
| apply RR at inference on SimpleShot, LaplacianShot, TIM | SimpleShot, LaplacianShot, TIM · mini-ImageNet, tiered-ImageNet, CUB (1-shot and 5-shot) | accuracy: SimpleShot+RR gains ~7-9% in 1-shot; TIM+RR >2% in all 1-shot tests; 5-shot gains marginal | Improved |
| apply RR at inference on SimpleShot, LaplacianShot, TIM | SimpleShot, LaplacianShot, TIM · mini-ImageNet, tiered-ImageNet, CUB 5-shot | accuracy: marginal gains (e.g., TIM+RR 85.04 to 85.20 on mini-ImageNet ResNet-18) | No clear effect |
| remove query-to-query propagation (alpha1=0) (ablation) | ProtoNet+MT, Matching+MT, Relation+MT, SimpleShot · mini-ImageNet 5-way 1-shot | accuracy: ProtoNet+MT 59.06 -> 57.86 when removed | Improved |
| remove query-to-support propagation (alpha2=0) (ablation) | ProtoNet+MT, Matching+MT, Relation+MT, SimpleShot · mini-ImageNet 5-way 1-shot | accuracy: ProtoNet+MT 59.06 -> 55.09 when removed; SimpleShot+RR 70.25 -> 64.14 | Improved |
| remove SSL loss during meta-training (ablation) | ProtoNet+MT · mini-ImageNet 5-way 1-shot | accuracy: 59.06 -> 58.85 when removed (marginal) | Improved |
| remove query-to-support propagation (alpha2=0) (ablation) | LaplacianShot+RR and TIM+RR · mini-ImageNet 5-way 1-shot | accuracy: LaplacianShot+RR 75.04 -> 72.85 (alpha2=0 in 1-shot); TIM+RR 76.54 -> 76.00 | Mixed |
| remove query-to-query propagation (alpha1=0) (ablation) | TIM+RR · mini-ImageNet 5-way 1-shot | accuracy: TIM+RR 76.54 -> 74.93 | Improved |
| apply multiple RR iterations at inference (beyond one) (ablation) | TIM and SimpleShot · mini-ImageNet, tiered-ImageNet, CUB 1-shot and 5-shot | accuracy | Hurt |
| increase number of query shots (more unlabeled queries) with RR | TIM+RR, SimpleShot+RR · 5-way 1-shot, mini-ImageNet and tiered-ImageNet | accuracy | Improved |
| apply RR with increasing support shots | TIM+RR, SimpleShot+RR · 5-way, varying support shots | accuracy | Mixed |
Closest papers by meaning
- Pushing the Limit of Fine-Tuning for Few-Shot Learning: Where Feature Reusing Meets Cross-Scale Attention (AAAI 2024 · similarity 0.89)
- EASE: Unsupervised Discriminant Subspace Learning for Transductive Few-Shot Learning (CVPR 2022 · similarity 0.88)
- TAFSSL: Task-Adaptive Feature Sub-Space Learning for few-shot classification (ECCV 2020 · similarity 0.88)
- Transductive Episodic-Wise Adaptive Metric for Few-Shot Learning (ICCV 2019 · similarity 0.88)
- Few-Shot Recognition via Stage-Wise Retrieval-Augmented Finetuning (CVPR 2025 · similarity 0.86)
- Strong Baselines for Parameter-Efficient Few-Shot Fine-Tuning (AAAI 2024 · similarity 0.86)
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.
