Research map / Graph learning / GNN architectures
ComENet: Towards Complete and Efficient Message Passing for 3D Molecular Graphs
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 |
|---|---|---|---|
| add rotation angles tau (global completeness) to ComENet (ablation) | ComENet · OC20 IS2RE average | energy MAE (eV): 0.6049 -> 0.5888 | Helped |
| add rotation angles tau (ablation) | ComENet · OC20 IS2RE OOD Both | energy MAE (eV): 0.6186 -> 0.5901 | Helped |
| add rotation angles tau (ablation) | ComENet · OC20 IS2RE ID | energy MAE (eV): 0.5585 -> 0.5558 | No clear effect |
| add rotation angles tau (ablation) | ComENet · OC20 IS2RE average | EwT: 3.42% -> 3.56% | Helped |
| use 1-hop ComENet message passing instead of 2-hop SphereNet | OC20 IS2RE · OC20 | training time per epoch: 290min -> 20min | Helped |
| use ComENet instead of SphereNet | Molecule3D · Molecule3D random split | HOMO-LUMO gap MAE: 0.0301 -> 0.0326 | Hurt |
| use ComENet instead of SphereNet | OC20 IS2RE · OC20 average | energy MAE (eV): 0.6023 -> 0.5888 | Helped |
| use ComENet instead of SphereNet | QM9 · QM9 | std. MAE (%): 0.91 -> 0.93 | No clear effect |
| use ComENet instead of SchNet | Molecule3D random split · Molecule3D | HOMO-LUMO gap MAE: 0.0428 -> 0.0326 | Helped |
Closest papers by meaning
- Empowering Active Learning for 3D Molecular Graphs with Geometric Graph Isomorphism (NeurIPS 2024 · similarity 0.87)
- RISE: Radius of Influence based Subgraph Extraction for 3D Molecular Graph Explanation (ICML 2025 · similarity 0.87)
- Lift Your Molecules: Molecular Graph Generation in Latent Euclidean Space (ICLR 2025 · similarity 0.86)
- HAGO-Net: Hierarchical Geometric Message Passing for Molecular Representation Learning (AAAI 2024 · similarity 0.85)
- On the Expressive Power of Sparse Geometric MPNNs (ICLR 2025 · similarity 0.85)
- Triplet Interaction Improves Graph Transformers: Accurate Molecular Graph Learning with Triplet Graph Transformers (ICML 2024 · similarity 0.85)
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.