Research map / Point clouds / Point cloud classification
Symmetry-adapted generation of 3d point sets for the targeted discovery of molecules
Niklas Gebauer, Michael Gastegger, Kristof Schütt
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 |
|---|---|---|---|
| Fine-tune on molecules with HOMO-LUMO gap ≤ 4.5 eV | G-SchNet · QM9 | molecules with HOMO-LUMO gap smaller than 5.0 eV: 7% -> 43% | Improved |
| Add the origin token (ablation) | G-SchNet · QM9 | molecules with 9 or fewer heavy atoms: 13.7% -> 60.4% | Improved |
| Add the origin token (ablation) | G-SchNet · QM9 | validity: drops by almost 20 percent without the origin token | Improved |
| Add the origin token (ablation) | G-SchNet · QM9 | ring-count distribution match | No clear effect |
| Train on QM9 excluding structures with 3- or 4-membered rings (ablation) | G-SchNet · QM9 subset without 3- and 4-membered rings | generated ring-count distribution | Improved |
| Increase sampling temperature to 1 or 2 (ablation) | G-SchNet · QM9 generated molecules | RMSD to relaxed structure | Hurt |
| Decrease sampling temperature below 0.1 (ablation) | G-SchNet · QM9 generated molecules | RMSD to relaxed structure | No clear effect |
| Use graph-based CGVAE instead of 3D G-SchNet generation | QM9 · QM9 | valid molecules: 77.07% -> 100.00% | Improved |
| Use graph-based CGVAE instead of 3D G-SchNet generation | QM9 · QM9 | novel molecules: 87.47% -> 94.30% | Improved |
Closest papers by meaning
- Lift Your Molecules: Molecular Graph Generation in Latent Euclidean Space (ICLR 2025 · similarity 0.87)
- WyckoffDiff -- A Generative Diffusion Model for Crystal Symmetry (ICML 2025 · similarity 0.87)
- 3D Molecule Generation from Rigid Motifs via $\mathrm{SE}(3)$ Flows (NeurIPS 2026 · similarity 0.87)
- Navigating the Design Space of Equivariant Diffusion-Based Generative Models for De Novo 3D Molecule Generation (ICLR 2024 · similarity 0.87)
- Symphony: Symmetry-Equivariant Point-Centered Spherical Harmonics for 3D Molecule Generation (ICLR 2024 · similarity 0.87)
- NEAT: Neighborhood-Guided, Efficient, Autoregressive Set Transformer for 3D Molecular Generation (NeurIPS 2026 · 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.
