Research map / Graph learning / GNN architectures
Can You Hear Me Now? A Benchmark for Long-Range Graph Propagation
Luca Miglior, Matteo Tolloso, Alessio Gravina, Davide Bacciu
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 |
|---|---|---|---|
| use global-attention transformer (GRIT) instead of local message passing (GCN) | GNN on ECHO-Synth · ECHO-Synth sssp | test MAE: 2.102 -> 0.121 | Improved |
| use GRIT instead of GCN | GNN on ECHO-Synth · ECHO-Synth diam | test MAE: 3.832 -> 1.014 | Improved |
| use multi-hop rewiring (DRew) instead of GCN | GNN on ECHO-Synth · ECHO-Synth ecc | test MAE: 5.233 -> 4.651 | Improved |
| use non-dissipative dynamics (A-DGN) instead of GraphCON | DE-GNN on ECHO-Synth · ECHO-Synth diam | test MAE: 2.969 -> 1.151 | Improved |
| use GPS instead of GCN | GNN on ECHO-Energy · ECHO-Energy | test MAE: 28.112 -> 5.257 | Improved |
| use SWAN instead of GIN | GNN on ECHO-Charge · ECHO-Charge (x1e-3) | test MAE: 10.784 -> 6.109 | Improved |
| use GRIT instead of GPS | graph transformer on ECHO-Energy · ECHO-Energy | test MAE: 5.257 -> 25.508 | Hurt |
| use GPS instead of GRIT | graph transformer on ECHO-Synth · ECHO-Synth sssp | test MAE: 0.121 -> 0.472 | Hurt |
| increase network depth (ablation) | GNNs on ECHO · ECHO | test MAE | Improved |
| vary readout depth (ablation) | GNNs on ECHO · ECHO | test MAE | No clear effect |
Closest papers by meaning
- Neural Atoms: Propagating Long-range Interaction in Molecular Graphs through Efficient Communication Channel (ICLR 2024 · similarity 0.88)
- Return of ChebNet: Understanding and Improving an Overlooked GNN on Long Range Tasks (NeurIPS 2025 · similarity 0.87)
- Long-Short-Range Message-Passing: A Physics-Informed Framework to Capture Non-Local Interaction for Scalable Molecular Dynamics Simulation (ICLR 2024 · similarity 0.87)
- On Measuring Long-Range Interactions in Graph Neural Networks (ICML 2025 · similarity 0.87)
- Adaptive Message Passing: A General Framework to Mitigate Oversmoothing, Oversquashing, and Underreaching (ICML 2025 · similarity 0.87)
- On the Scalability of GNNs for Molecular Graphs (NeurIPS 2024 · 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.
