Research map / Probabilistic and scientific modeling / Molecules and proteins
Towards Combinatorial Generalization for Catalysts: A Kohn-Sham Charge-Density Approach
Phil Pope, David Jacobs
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 |
|---|---|---|---|
| initialize SCF with learned density instead of atomic charge superposition | DFT SCF cycle · Test (binary) | % structures with positive s-AUC savings: 83 | Improved |
| initialize SCF with learned density instead of atomic charge superposition | DFT SCF cycle · Test (ternary) | % structures with positive s-AUC savings: 86 | Improved |
| initialize SCF with learned density instead of atomic charge superposition | DFT SCF cycle · Test (binary) | mean iterations saved (%): 13 | Improved |
| initialize SCF with learned density instead of atomic charge superposition | DFT SCF cycle · Test (ternary) | mean iterations saved (%): 13 | Improved |
| initialize SCF with learned density | DFT SCF cycle · Test (binary) | min iterations saved (%): -53 | Hurt |
| initialize SCF with learned density | DFT SCF cycle · Test (ternary) | min iterations saved (%): -90 | Hurt |
| test on unseen element combinations vs training structures | SCN density model · Binary | mean iterations saved (%): 22 -> 13 | Mixed |
| train on unary and binary, test on ternary | SCN density model · Test (ternary) | % structures with positive s-AUC savings: 86 | Improved |
| use SCN instead of DimeNet++ or GemNet | density model · initial experiments | density prediction quality | Improved |
| use learned density as SCF start | DFT converged energy · Test set | fraction with zero energy difference: 88 | No clear effect |
| use learned density as SCF start | DFT band gap · Test set | fraction exactly matching band gap: 95 | No clear effect |
Closest papers by meaning
- Enhancing the Scalability and Applicability of Kohn-Sham Hamiltonians for Molecular Systems (ICLR 2025 · similarity 0.88)
- A Recipe for Charge Density Prediction (NeurIPS 2024 · similarity 0.88)
- EDBench: Large-Scale Electron Density Data for Molecular Modeling (NeurIPS 2025 · similarity 0.87)
- ELECTRA: A Cartesian Network for 3D Charge Density Prediction with Floating Orbitals (NeurIPS 2025 · similarity 0.86)
- Learning the Electronic Hamiltonian of Large Atomic Structures (ICML 2025 · similarity 0.86)
- Gaussian Plane-Wave Neural Operator for Electron Density Estimation (ICML 2024 · 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.
