Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
ECKPN: Explicit Class Knowledge Propagation Network for Transductive Few-Shot Learning
Check 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 the calibration module to squeezed class knowledge (ablation) | ECKPN · miniImageNet 5-way 1-shot, Conv-4 | accuracy: 68.24 -> 68.89 | Helped |
| add the calibration module to squeezed class knowledge (ablation) | ECKPN · miniImageNet 5-way 1-shot, ResNet-12 | accuracy: 69.76 -> 70.48 | Helped |
| add squeeze and calibration modules (ablation) | ECKPN · miniImageNet 5-way 1-shot, Conv-4 | accuracy: 67.32 -> 68.89 | Helped |
| add squeeze and calibration modules (ablation) | ECKPN · miniImageNet 5-way 1-shot, ResNet-12 | accuracy: 69.12 -> 70.48 | Helped |
| add semantic class knowledge (ablation) | ECKPN calibration module · miniImageNet 5-way 1-shot, Conv-4 | accuracy: 68.53 -> 68.89 | Helped |
| add semantic class knowledge (ablation) | ECKPN calibration module · miniImageNet 5-way 1-shot, ResNet-12 | accuracy: 69.97 -> 70.48 | Helped |
| add visual class knowledge (ablation) | ECKPN calibration module · miniImageNet 5-way 1-shot, Conv-4 | accuracy: 68.16 -> 68.89 | Helped |
| add visual class knowledge (ablation) | ECKPN calibration module · miniImageNet 5-way 1-shot, ResNet-12 | accuracy: 69.61 -> 70.48 | Helped |
| increase the number of comparison message-passing layers (ablation) | ECKPN · miniImageNet 5-way 1-shot | accuracy | Mixed |
| use multi-head sample relations instead of a single relation head (ablation) | ECKPN comparison module · miniImageNet 5-way 1-shot | accuracy | Mixed |
Closest papers by meaning
- LEARNING TO PROPAGATE LABELS: TRANSDUCTIVE PROPAGATION NETWORK FOR FEW-SHOT LEARNING (ICLR 2019 · similarity 0.90)
- Knowledge Graph Transfer Network for Few-Shot Recognition (AAAI 2020 · similarity 0.87)
- Learning to Compare: Relation Network for Few-Shot Learning (CVPR 2018 · similarity 0.87)
- Graph Few-Shot Learning via Knowledge Transfer (AAAI 2020 · similarity 0.86)
- Simple Semantic-Aided Few-Shot Learning (CVPR 2024 · similarity 0.85)
- Envisioning Class Entity Reasoning by Large Language Models for Few-shot Learning (AAAI 2025 · similarity 0.85)
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.