Research map / Probabilistic and scientific modeling / Molecules and proteins
Take Note: Your Molecular Dataset Is Probably Aligned
Peter Lippmann, Roman Remme, Manuel V. Klockow, Fred A. Hamprecht
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 on canonical orientation instead of random rotation | orientation-only MLP · QM9 epsilon_LUMO | test MSE: 1.6367 -> 1.4237 | Improved |
| train on canonical orientation instead of random rotation | orientation-only MLP · QM9 ZPVE | test MSE: 0.8111 -> 0.6204 | Improved |
| train on canonical orientation instead of random rotation | orientation-only MLP · QM9 cV | test MSE: 16.173 -> 13.814 | Improved |
| train on canonical orientation instead of random rotation | orientation-only MLP · QMugs URT | test MSE: 890.55 -> 843.48 | Improved |
| train on canonical orientation instead of random rotation | orientation-only MLP · OMol25 Etot | test MSE: 14,398.7 × 106 -> 13,689.1 × 106 | Improved |
| randomly rotate the data | orientation-only MLP · QMugs V_ee | test MSE: (4,596.6 ± 0.7) × 103 -> (4,894.6 ± 0.2) × 103 | Hurt |
| add Gaussian noise up to 1 Angstrom to atom positions | message passing orientation detector · QM9, QMugs, OMol25 | test accuracy | No clear effect |
Closest papers by meaning
- Learning Over Molecular Conformer Ensembles: Datasets and Benchmarks (ICLR 2024 · similarity 0.86)
- Physical Consistency Bridges Heterogeneous Data in Molecular Multi-Task Learning (NeurIPS 2024 · similarity 0.85)
- Learning Molecular Representation in a Cell (ICLR 2025 · similarity 0.85)
- Atomas: Hierarchical Adaptive Alignment on Molecule-Text for Unified Molecule Understanding and Generation (ICLR 2025 · similarity 0.85)
- Towards Foundational Models for Molecular Learning on Large-Scale Multi-Task Datasets (ICLR 2024 · similarity 0.85)
- TRIDENT: Tri-Modal Molecular Representation Learning with Taxonomic Annotations and Local Correspondence (NeurIPS 2025 · 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.
