Research map / Graph learning / GNN architectures
Graph Few-shot Learning with Task-specific Structures
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 |
|---|---|---|---|
| GLITTER vs best baseline G-Meta | GLITTER · Fold-PPI shared graphs, disjoint labels | accuracy (%): 46.1 -> 53.3 | Helped |
| GLITTER vs best baseline GPN | GLITTER · DBLP 5-way 3-shot | accuracy (%): 75.4 -> 79.3 | Helped |
| GLITTER vs best baseline RALE | GLITTER · DBLP 10-way 3-shot | accuracy (%): 65.1 -> 69.5 | Helped |
| GLITTER vs best baseline G-Meta | GLITTER · ogbn-arxiv 10-way 3-shot | accuracy (%): 38.4 -> 44.2 | Helped |
| add relevant node sampling (local and common sampling) (ablation) | GLITTER · Tissue-PPI/Fold-PPI/single graph | accuracy (%) | Helped |
| add node influence enhancement within classes (ablation) | GLITTER · single and multiple graph settings | accuracy (%) | Helped |
| add mutual information maximization on query nodes (ablation) | GLITTER · single and multiple graph settings | accuracy (%) | Helped |
| disjoint graphs with disjoint labels vs shared graphs and labels | GLITTER · Tissue-PPI | accuracy (%): 69.7 -> 60.3 | Hurt |
| increase N from 5 to 10 ways | GLITTER · DBLP | accuracy (%): 79.3 -> 69.5 | Hurt |
Closest papers by meaning
- Cross-Domain Few-Shot Graph Classification (AAAI 2022 · similarity 0.91)
- Graph Few-Shot Learning via Knowledge Transfer (AAAI 2020 · similarity 0.91)
- Unlocking the Potential of Black-box Pre-trained GNNs for Graph Few-shot Learning (AAAI 2025 · similarity 0.89)
- Graph Few-Shot Learning via Adaptive Spectrum Experts and Cross-Set Distribution Calibration (NeurIPS 2025 · similarity 0.89)
- Fast Graph Sharpness-Aware Minimization for Enhancing and Accelerating Few-Shot Node Classification (NeurIPS 2024 · similarity 0.87)
- LLMs as Zero-shot Graph Learners: Alignment of GNN Representations with LLM Token Embeddings (NeurIPS 2024 · 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.