Research map / Point clouds / Point cloud classification
Adaptive Canonicalization with Application to Invariant Anisotropic Geometric Networks
Ya-Wei Eileen Lin, Ron Levie
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 adaptive canonicalization to NLSF (A-NLSF) | NLSF · Grid signal orientation | accuracy: 50.07 -> 99.38 | Improved |
| add adaptive canonicalization to NLSF | NLSF · TUDataset ENZYMES | accuracy: 65.94 -> 73.01 | Improved |
| add adaptive canonicalization to NLSF | NLSF · TUDataset MUTAG | accuracy: 84.13 -> 87.94 | Improved |
| use adaptive canonicalization instead of OAP canonicalization | GIN · TUDataset PROTEINS | accuracy: 83.41 -> 85.47 | Improved |
| add fixed canonicalization (OAP) to GIN | GIN · TUDataset ENZYMES | accuracy: 45.69 -> 58.40 | Improved |
| add fixed canonicalization (OAP) to GIN | GIN · Grid signal orientation | accuracy: 50.00 -> 50.03 | No clear effect |
| add frame averaging to GIN | GIN · TUDataset PROTEINS | accuracy: 76.02 -> 79.53 | Improved |
| use A-NLSF instead of S2GNN | S2GNN · TUDataset NCI1 | accuracy: 75.62 -> 82.01 | Improved |
| use A-NLSF instead of GPS | GPS · ogbg-molhiv | AUROC: 0.7880 -> 0.8019 | Improved |
| use A-NLSF instead of OAP canonicalization | GatedGCN · ogbg-ppa | accuracy: 0.7745 -> 0.8149 | Improved |
| add adaptive canonicalization to DGCNN | DGCNN · ModelNet40 | accuracy: 89.0 -> 91.6 | Improved |
| add adaptive canonicalization to PointNet | PointNet · ModelNet40 | accuracy: 75.8 -> 81.1 | Improved |
Closest papers by meaning
- A Canonicalization Perspective on Invariant and Equivariant Learning (NeurIPS 2024 · similarity 0.90)
- On the Generalization of Equivariant Graph Neural Networks (ICML 2024 · similarity 0.88)
- Learning (Approximately) Equivariant Networks via Constrained Optimization (NeurIPS 2025 · similarity 0.88)
- Improving Equivariant Networks with Probabilistic Symmetry Breaking (ICLR 2025 · similarity 0.87)
- Approximately Piecewise E(3) Equivariant Point Networks (ICLR 2024 · similarity 0.86)
- Universally Invariant Learning in Equivariant GNNs (NeurIPS 2025 · 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.
