Research map / Graph learning / GNN architectures
Node Feature Extraction by Self-Supervised Multi-scale Neighborhood Prediction
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 GIANT-XRT features instead of OGB features | RevGAT+SelfKD · ogbn-arxiv | accuracy: 74.26 ± 0.17 -> 76.12 ± 0.16 | Helped |
| Use GIANT-XRT features instead of OGB features | SAGN+SLE · ogbn-products | accuracy: 84.28 ± 0.14 -> 85.47 ± 0.29 | Helped |
| Use GIANT-XRT features instead of OGB features | GAMLP+RLU · ogbn-papers100M | accuracy: 68.25 ± 0.19 -> 69.67 ± 0.04 | Helped |
| Use GIANT-XRT features instead of OGB features | MLP · ogbn-papers100M | accuracy: 47.24 ± 0.31 -> 61.10 ± 0.19 | Helped |
| Replace neighborhood prediction with link prediction (ablation) | GIANT-XRT feature extraction · ogbn-arxiv and ogbn-products | accuracy: 73.08 ± 0.06 -> 67.33 ± 0.54; 79.82 ± 0.07 -> 73.83 ± 0.06 | Mixed |
| Add TFIDF features to hierarchical clustering (ablation) | GIANT-XRT · ogbn-arxiv and ogbn-products | accuracy: 72.74 ± 0.17 -> 73.08 ± 0.06; 78.91 ± 0.28 -> 79.82 ± 0.07 | Helped |
| Add PIFA embeddings to hierarchical clustering (ablation) | GIANT-XRT · ogbn-arxiv and ogbn-products | accuracy: 71.74 ± 0.15 -> 73.08 ± 0.06; 79.37 ± 0.15 -> 79.82 ± 0.07 | Helped |
| Use both TFIDF and PIFA for clustering instead of either alone (ablation) | GIANT-XRT · ogbn-arxiv and ogbn-products | accuracy: 72.74 ± 0.17 -> 73.08 ± 0.06; 83.83 ± 0.14 -> 84.40 ± 0.17 | Helped |
| Remove both TFIDF and PIFA from hierarchical clustering (ablation) | GIANT-XRT · ogbn-arxiv and ogbn-products | accuracy: 73.08 ± 0.06 -> 69.33 ± 0.19; 79.82 ± 0.07 -> 74.16 ± 0.22 | Hurt |
Closest papers by meaning
- Label-free Node Classification on Graphs with Large Language Models (LLMs) (ICLR 2024 · similarity 0.90)
- Beyond Real-world Benchmark Datasets: An Empirical Study of Node Classification with GNNs (NeurIPS 2022 · similarity 0.90)
- NLGT: Neighborhood-based and Label-enhanced Graph Transformer Framework for Node Classification (AAAI 2025 · similarity 0.90)
- Subgraph Invariant Learning Towards Large-Scale Graph Node Classification (AAAI 2025 · similarity 0.89)
- Node Identifiers: Compact, Discrete Representations for Efficient Graph Learning (ICLR 2025 · similarity 0.89)
- A Topological Perspective on Demystifying GNN-Based Link Prediction Performance (ICLR 2024 · similarity 0.88)
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.