Research map / Graph learning / GNN architectures
On the Equivalence Between Temporal and Static Equivariant Graph Representations
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 |
|---|---|---|---|
| Time-then-graph (GRU-GCN) instead of time-and-graph baselines (1-WL GNN components) | temporal graph representation · DynCSL | ROCAUC: 1.00 vs ~0.50-0.52 for baselines | Helped |
| Time-then-graph (GRU-GCN) vs TGAT/TGN time-then-graph variants that only use observed edges | temporal graph representation · DynCSL | ROCAUC: 1.00 vs 0.48 (TGAT), 0.51 (TGN) | Helped |
| Time-then-graph (GRU-GCN) vs time-and-graph baselines | temporal graph representation · Brain-10 | ROCAUC: 0.91 vs 0.84 (DCRNN), 0.87 (GCN-GRU) | Helped |
| Time-then-graph (GRU-GCN) on temporal node regression | temporal graph representation · PeMS04, PeMS08, Spain-COVID, England-COVID (MAPE) | MAPE (%): best on PeMS08 transductive and England-COVID; second best on PeMS04 and Spain-COVID transductive | Mixed |
| Time-then-graph (GRU-GCN) vs time-and-graph baselines in training speed and memory | temporal graph representation · GPU efficiency on PeMS04, PeMS08, Spain-COVID, England-COVID | training time per minibatch and peak GPU memory: 7 ms vs 17 ms (GCN-GRU) and 25 ms (TGN) on PeMS04 | Helped |
Closest papers by meaning
- TGM: A Modular and Efficient Library for Machine Learning on Temporal Graphs (ICLR 2026 · similarity 0.90)
- A Deep Probabilistic Framework for Continuous Time Dynamic Graph Generation (AAAI 2025 · similarity 0.88)
- Using Time-Aware Graph Neural Networks to Predict Temporal Centralities in Dynamic Graphs (NeurIPS 2024 · similarity 0.88)
- Temporal Generalization Estimation in Evolving Graphs (ICLR 2024 · similarity 0.87)
- Simple and Efficient Heterogeneous Temporal Graph Neural Network (NeurIPS 2025 · similarity 0.87)
- Equivariance Everywhere All At Once: A Recipe for Graph Foundation Models (NeurIPS 2025 · 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.