Research map / Graph learning / GNN architectures
GLEMOS: Benchmark for Instantaneous Graph Learning Model Selection
Namyong Park, Ryan A. Rossi, Xing Wang, Antoine Simoulin, Nesreen K. Ahmed, Christos Faloutsos
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| increase sparsity of performance matrix from 10% to 90% | AS · Sparse LP | AUC: 0.829 -> 0.547 | Hurt |
| increase sparsity from 10% to 90% | MetaGL · Sparse LP | AUC: 0.865 -> 0.811 | Hurt |
| increase sparsity from 10% to 90% | S2 · Sparse LP | AUC: 0.813 -> 0.464 | Hurt |
| increase sparsity from 10% to 90% | GB-Rank · Sparse NC | AUC: 0.744 -> 0.697 | Hurt |
| increase sparsity from 10% to 90% | MetaGL · Sparse NC | AUC: 0.761 -> 0.659 | Hurt |
| use meta-graph features vs only prior performance (GB-Perf) | AS · Fully-Observed LP | AUC: 0.735 -> 0.864 | Improved |
| use optimizable method MetaGL instead of random selection | MetaGL · Fully-Observed LP | AUC: 0.524 -> 0.875 | Improved |
| optimizable method MetaGL vs simple AS on small node classification set | MetaGL · Fully-Observed NC | AUC: 0.762 -> 0.740 | Hurt |
| evaluate on out-of-domain graphs instead of in-domain | AS · Out-Of-Domain LP | AUC: 0.864 -> 0.786 | Hurt |
| evaluate on out-of-domain graphs instead of in-domain | MetaGL · Out-Of-Domain LP | AUC: 0.875 -> 0.871 | No clear effect |
| train on small graphs, test on large graphs | MetaGL · Small-To-Large LP | AUC: 0.875 -> 0.870 | No clear effect |
| train on small graphs, test on large graphs | S2 · Small-To-Large LP | AUC: 0.809 -> 0.767 | Hurt |
Closest papers by meaning
- GLBench: A Comprehensive Benchmark for Graph with Large Language Models (NeurIPS 2024 · similarity 0.89)
- Classic GNNs are Strong Baselines: Reassessing GNNs for Node Classification (NeurIPS 2024 · similarity 0.87)
- GC4NC: A Benchmark Framework for Graph Condensation on Node Classification with New Insights (NeurIPS 2025 · similarity 0.86)
- Can Classic GNNs Be Strong Baselines for Graph-level Tasks? Simple Architectures Meet Excellence (ICML 2025 · similarity 0.86)
- No Metric to Rule Them All: Toward Principled Evaluations of Graph-Learning Datasets (ICML 2025 · similarity 0.86)
- Diffusing to the Top: Boost Graph Neural Networks with Minimal Hyperparameter Tuning (ICLR 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.
