Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Looking Wider for Better Adaptive Representation in Few-Shot Learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use CNL with task-specific scaling and MI constraint instead of DeepEMD | ResNet12 few-shot classifier · MiniImageNet 5-way 1-shot | accuracy: 65.91 ± 0.82 -> 67.96 ± 0.98 | Helped |
| Use CNL with task-specific scaling and MI constraint instead of DeepEMD | ResNet12 few-shot classifier · MiniImageNet 5-way 5-shot | accuracy: 82.41 ± 0.56 -> 83.36 ± 0.51 | Helped |
| Use CNL with task-specific scaling and MI constraint instead of DeepEMD | ResNet12 few-shot classifier · TieredImageNet 5-way 1-shot | accuracy: 71.16 ± 0.87 -> 73.42 ± 0.95 | Helped |
| Use CNL with task-specific scaling and MI constraint instead of DeepEMD | ResNet12 few-shot classifier · TieredImageNet 5-way 5-shot | accuracy: 86.03 ± 0.58 -> 87.72 ± 0.75 | Helped |
| Add task-specific scaling to local features (ablation) | ResNet12 local-feature classifier · MiniImageNet 5-way 1-shot | accuracy: 63.08 ± 0.62 -> 66.61 ± 0.85 | Helped |
| Add task-specific scaling to local features (ablation) | ResNet12 local-feature classifier · TieredImageNet 5-way 1-shot | accuracy: 69.34 ± 0.81 -> 70.97 ± 0.85 | Helped |
| Add non-local global features, concatenated with local features (ablation) | ResNet12 few-shot classifier · MiniImageNet 5-way 1-shot | accuracy: 63.08 ± 0.62 -> 64.20 ± 0.98 | Helped |
| Add non-local global features, concatenated with local features (ablation) | ResNet12 few-shot classifier · TieredImageNet 5-way 1-shot | accuracy: 69.34 ± 0.81 -> 71.21 ± 0.85 | Helped |
| Use CNL task-specific features instead of local features alone (ablation) | ResNet12 few-shot classifier · MiniImageNet 5-way 1-shot | accuracy: 63.08 ± 0.62 -> 64.97 ± 0.98 | Helped |
| Use CNL task-specific features instead of local features alone (ablation) | ResNet12 few-shot classifier · TieredImageNet 5-way 1-shot | accuracy: 69.34 ± 0.81 -> 72.02 ± 0.85 | Helped |
| Add task-specific scaling to CNL task-specific features (ablation) | ResNet12 CNL classifier · MiniImageNet 5-way 1-shot | accuracy: 64.97 ± 0.98 -> 67.83 ± 0.99 | Helped |
| Add task-specific scaling to CNL task-specific features (ablation) | ResNet12 CNL classifier · TieredImageNet 5-way 1-shot | accuracy: 72.02 ± 0.85 -> 72.98 ± 0.85 | Helped |
Closest papers by meaning
- Cross-Domain Few-Shot Learning With Task-Specific Adapters (CVPR 2022 · similarity 0.89)
- Dense Classification and Implanting for Few-Shot Learning (CVPR 2019 · similarity 0.89)
- TADAM: Task dependent adaptive metric for improved few-shot learning (NeurIPS 2018 · similarity 0.89)
- Pushing the Limit of Fine-Tuning for Few-Shot Learning: Where Feature Reusing Meets Cross-Scale Attention (AAAI 2024 · similarity 0.88)
- Discriminative Sample-Guided and Parameter-Efficient Feature Space Adaptation for Cross-Domain Few-Shot Learning (CVPR 2024 · similarity 0.88)
- Making Large Vision Language Models to Be Good Few-Shot Learners (AAAI 2025 · similarity 0.87)
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.