Research map / Image generation / Fast diffusion sampling
3D molecule generation by denoising voxel grids
Pedro O. Pinheiro, Joshua Rackers, joseph Kleinhenz, Michael Maser, Omar Mahmood, Andrew Martin Watkins, Stephen Ra, Vishnu Sresht, Saeed Saremi
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 |
|---|---|---|---|
| train with random rotation and translation augmentation (ablation) | VoxMol · QM9 | stable molecules %: 84.2 -> 89.3 | Improved |
| train with augmentation (ablation) | VoxMol · QM9 | unique %: 77.2 -> 92.1 | Improved |
| train with augmentation (ablation) | VoxMol · QM9 | bond angle W1: 3.80 -> 1.96 | Improved |
| train with augmentation (ablation) | VoxMol · GEOM-drugs | stable molecules %: 44.4 -> 75.0 | Improved |
| train with augmentation (ablation) | VoxMol · GEOM-drugs | bond angle W1: 2.14 -> 0.64 | Improved |
| replace EDM with VoxMol | VoxMol · GEOM-drugs | stable molecules %: 40.3 -> 75.0 | Improved |
| replace EDM with VoxMol | VoxMol · QM9 | stable molecules %: 97.9 -> 89.3 | Hurt |
| increase walk steps from 50 to 1,000 (ablation) | VoxMol · GEOM-drugs | unique %: 87.8 -> 99.8 | Improved |
| increase walk steps from 50 to 1,000 (ablation) | VoxMol · GEOM-drugs | stable molecules %: 78.9 -> 75.5 | Hurt |
| increase walk steps from 50 to 1,000 (ablation) | VoxMol · GEOM-drugs | avg time s/mol: 0.90 -> 14.9 | Hurt |
| add coordinate refinement (ablation) | VoxMol · QM9 | stable molecules %: 80.5 -> 89.3 | Improved |
| add coordinate refinement (ablation) | VoxMol · GEOM-drugs | bond angle W1: 2.92 -> .63 | Improved |
Closest papers by meaning
- Structure-based drug design by denoising voxel grids (ICML 2024 · similarity 0.93)
- Score-based 3D molecule generation with neural fields (NeurIPS 2024 · similarity 0.91)
- Geometric-Facilitated Denoising Diffusion Model for 3D Molecule Generation (AAAI 2024 · similarity 0.90)
- Towards Unified and Lossless Latent Space for 3D Molecular Latent Diffusion Modeling (NeurIPS 2025 · similarity 0.88)
- Scalable Non-Equivariant 3D Molecule Generation via Rotational Alignment (ICML 2025 · similarity 0.87)
- Lift Your Molecules: Molecular Graph Generation in Latent Euclidean Space (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.
