Research map / Graph learning / GNN architectures
Facilitating Graph Neural Networks with Random Walk on Simplicial Complexes
Cai Zhou, Xiyuan Wang, Muhan Zhang
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 |
|---|---|---|---|
| add undirected EdgeRWSE (ablation) | GAT · ZINC-12k test | MAE: 0.384 -> 0.163 | Improved |
| add Hodge1Lap project (ablation) | GAT · ZINC-12k test | MAE: 0.384 -> 0.130 | Improved |
| add Hodge1Lap project (ablation) | PNA · ZINC-12k test | MAE: 0.188 -> 0.074 | Improved |
| add RWSE (ablation) | GINE · ZINC-12k test | MAE: 0.133 -> 0.074 | Improved |
| add Hodge1Lap project on top of RWSE (ablation) | GINE · ZINC-12k test | MAE: 0.074 -> 0.068 | Improved |
| add RWMP on top of RWSE (ablation) | GINE · ZINC-12k test | MAE: 0.074 -> 0.068 | No clear effect |
| directed vs undirected EdgeRWSE (ablation) | GINE · ZINC-12k test | MAE: 0.110 -> 0.104 | No clear effect |
| Hodge1Lap project instead of abs (ablation) | GINE · ZINC-12k test | MAE: 0.102 -> 0.091 | Improved |
| replace RWSE by Inter-RWSE (ablation) | GINE · ZINC-12k test | MAE: 0.069 -> 0.083 | Hurt |
| add LapPE (ablation) | GINE · ZINC-12k test | MAE: 0.133 -> 0.120 | Improved |
| add RRWP plus Hodge1Lap project to GRIT (ablation) | GRIT · ZINC-12k test | MAE: 0.059 -> 0.057 | Improved |
| add RWSE and Hodge1Lap project (ablation) | GPS · ZINC-12k test | MAE: 0.113 -> 0.064 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Specformer — Specformer: Spectral Graph Neural Networks Meet Transformers (ICLR 2023) · better on 2 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- What Are Good Positional Encodings for Directed Graphs? (ICLR 2025 · similarity 0.88)
- Revisiting Random Walks for Learning on Graphs (ICLR 2025 · similarity 0.88)
- Learning From Simplicial Data Based on Random Walks and 1D Convolutions (ICLR 2024 · similarity 0.88)
- Random Search Neural Networks for Efficient and Expressive Graph Learning (NeurIPS 2025 · similarity 0.87)
- Learning Long Range Dependencies on Graphs via Random Walks (ICLR 2025 · similarity 0.86)
- Non-convolutional graph neural networks. (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.
