Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Laplacian Regularized 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 |
|---|---|---|---|
| Add query-query Laplacian regularization to nearest-prototype classification (ablation) | Few-shot classification with ResNet-18 · miniImageNet 1-shot | accuracy: 63.10 -> 66.20 | Helped |
| Add query-query Laplacian regularization to nearest-prototype classification (ablation) | Few-shot classification with ResNet-18 · tieredImageNet 1-shot | accuracy: 69.68 -> 72.89 | Helped |
| Add query-query Laplacian regularization to nearest-prototype classification (ablation) | Few-shot classification with ResNet-18 · CUB 1-shot | accuracy: 70.28 -> 74.46 | Helped |
| Use rectified prototypes instead of standard support prototypes (ablation) | LaplacianShot with ResNet-18 · miniImageNet 1-shot | accuracy: 63.10 -> 69.74 | Helped |
| Use rectified prototypes instead of standard support prototypes (ablation) | LaplacianShot with ResNet-18 · CUB 1-shot | accuracy: 70.28 -> 78.76 | Helped |
| Add Laplacian regularization when using rectified prototypes (ablation) | LaplacianShot with ResNet-18 · miniImageNet 1-shot | accuracy: 69.74 -> 72.11 | Helped |
| Use rectified prototypes with Laplacian regularization (ablation) | LaplacianShot on iNat · iNat | accuracy | Hurt |
| Use LaplacianShot instead of the strongest reported baseline | ResNet-18 few-shot classification · miniImageNet 1-shot | accuracy: 67.19 ± 0.55 -> 72.11 ± 0.19 | Helped |
| Use LaplacianShot instead of the strongest reported baseline | ResNet-18 few-shot classification · tieredImageNet 1-shot | accuracy: 73.21 ± 0.58 -> 78.98 ± 0.21 | Helped |
| Use LaplacianShot instead of the strongest reported baseline | ResNet-18 few-shot classification · CUB 1-shot | accuracy: 73.49 -> 80.96 | Helped |
| Use unnormalized rather than L2-normalized features | LaplacianShot with ResNet-18 · iNat Per Class | accuracy: 58.72 -> 62.80 | Helped |
Closest papers by meaning
- Realistic evaluation of transductive few-shot learning (NeurIPS 2021 · similarity 0.90)
- Information Maximization for Few-Shot Learning (NeurIPS 2020 · similarity 0.90)
- Few-Shot Segmentation Without Meta-Learning: A Good Transductive Inference Is All You Need? (CVPR 2021 · similarity 0.89)
- LEARNING TO PROPAGATE LABELS: TRANSDUCTIVE PROPAGATION NETWORK FOR FEW-SHOT LEARNING (ICLR 2019 · similarity 0.88)
- Transductive Zero-Shot and Few-Shot CLIP (CVPR 2024 · similarity 0.87)
- Graph Few-Shot Learning via Adaptive Spectrum Experts and Cross-Set Distribution Calibration (NeurIPS 2025 · 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.