Research map / Graph learning / GNN architectures
DRew: Dynamically Rewired Message Passing with Delay
Benjamin Gutteridge, Xiaowen Dong, Michael M. Bronstein, Francesco Di Giovanni
This paper tried a common change: Add positional encoding
Add Laplacian positional encoding · Peptides-func · AP: 0.6996 ±0.0076 -> 0.7150 ±0.0044 Improved
What 646 accepted papers report for this change
- 525 improved
- 58 no clear effect
- 7 mixed
- 56 hurt
- Learning-Augmented Scalable Linear Assignment Problem Optimization via Neural Dual Warm-Starts ICML 2026 · add positional encodings to features (d=13 -> d=21) · Block-structured, N=4096 · speedupImproved
- Mesh Based Simulations with Spatial and Temporal awareness ICML 2026 · use learned positional embedding, learnable relative bias or weighted adjacency instead of RoPE · Aneurysm · RMSEHurt
- Sparse ActionGen: Accelerating Diffusion Policy with Real-time Pruning ICML 2026 · add sinusoidal positional encoding to the pruner · Robomimic Tool Hang · success rate (%): 28 -> 54Improved
See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.
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 |
|---|---|---|---|
| Replace standard GCN message passing with layer-dependent dynamic rewiring | GCN · Peptides-func | AP: 0.5930 ±0.0023 -> 0.6996 ±0.0076 | Improved |
| Replace standard GCN message passing with layer-dependent dynamic rewiring | GCN · Peptides-struct | MAE: 0.3496 ±0.0013 -> 0.2781 ±0.0028 | Improved |
| Replace standard GCN message passing with layer-dependent dynamic rewiring | GCN · PCQM-Contact | MRR: 0.3234 ±0.0006 -> 0.3444 ±0.0017 | Improved |
| Replace GatedGCN message passing with layer-dependent dynamic rewiring | GatedGCN · PascalVOC-SP | F1: 0.2873 ±0.0219 -> 0.3214 ±0.0021 | Improved |
| Add Laplacian positional encoding (ablation) | DRew-GCN · Peptides-func | AP: 0.6996 ±0.0076 -> 0.7150 ±0.0044 | Improved |
| Add Laplacian positional encoding (ablation) | DRew-GCN · Peptides-struct | MAE: 0.2781 ±0.0028 -> 0.2536 ±0.0015 | Improved |
| Add Laplacian positional encoding (ablation) | DRew-GCN · PCQM-Contact | MRR: 0.3444 ±0.0017 -> 0.3442 ±0.0006 | No clear effect |
| Add Laplacian positional encoding (ablation) | DRew-GatedGCN · PascalVOC-SP | F1: 0.3214 ±0.0021 -> 0.3314 ±0.0024 | Improved |
| Use delayed propagation with ν=1 instead of no delay (ablation) | DRew-GIN · QM9 mu | MAE: 1.93 ±0.06 -> 2.00 ±0.05 | Hurt |
| Use delayed propagation with ν=1 instead of no delay (ablation) | DRew-GIN · QM9 alpha | MAE: 1.63 ±0.03 -> 1.63 ±0.05 | No clear effect |
| Use delayed propagation with ν=1 instead of no delay (ablation) | DRew-GIN · QM9 R2 | MAE: 9.39 ±0.13 -> 9.94 ±0.07 | Hurt |
Closest papers by meaning
- PANDA: Expanded Width-Aware Message Passing Beyond Rewiring (ICML 2024 · similarity 0.90)
- Probabilistic Graph Rewiring via Virtual Nodes (NeurIPS 2024 · similarity 0.90)
- Probabilistically Rewired Message-Passing Neural Networks (ICLR 2024 · similarity 0.89)
- Locality-Aware Graph Rewiring in GNNs (ICLR 2024 · similarity 0.88)
- Towards Dynamic Message Passing on Graphs (NeurIPS 2024 · similarity 0.87)
- The Effectiveness of Curvature-Based Rewiring and the Role of Hyperparameters in GNNs Revisited (ICLR 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.
