Research map / Probabilistic and scientific modeling / Dynamical systems
Orbital Transformers for Predicting Wavefunctions in Time-Dependent Density Functional Theory
Xuan Zhang, Haiyang Yu, Chengdong Wang, Jacob Helwig, Shuiwang Ji, Xiaofeng Qian
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 |
|---|---|---|---|
| Use density matrix instead of layer-wise pooling for state interaction | OrbEvo · QM9 | Rollout nRMSE: 0.2071 -> 0.1885 | Improved |
| Add pushforward training (ablation) | OrbEvo-DM-s8 · MDA | Rollout nRMSE: 0.2851 -> 0.1778 | Improved |
| Add pushforward training (ablation) | OrbEvo-WF-sall · MDA | Rollout nRMSE: 0.3708 -> 0.1585 | Improved |
| Add time-evolving density matrix dependency (ablation) | OrbEvo-DM-s8 · MDA | Rollout nRMSE: 0.5457 -> 0.1778 | Improved |
| Model electronic states as individual graphs rather than channel concatenation (ablation) | OrbEvo-DM · MDA | Rollout nRMSE: 0.7875 -> 0.1888 | Improved |
| Use equivariant electric field conditioning instead of invariant bias (ablation) | OrbEvo-WF-sall · MDA | Rollout nRMSE: 1.1564 -> 0.1585 | Improved |
| Sample 8 electronic states instead of all in training (ablation) | OrbEvo-DM · MDA | Rollout nRMSE: 0.1888 -> 0.1778 | No clear effect |
| Sample 4 electronic states instead of all in training (ablation) | OrbEvo-DM · MDA | Rollout nRMSE: 0.1888 -> 0.1902 | No clear effect |
| Sample 8 electronic states instead of all in training (ablation) | OrbEvo-WF · MDA | Rollout nRMSE: 0.1585 -> 0.4054 | Hurt |
| Sample 4 electronic states instead of all in training (ablation) | OrbEvo-WF · MDA | Rollout nRMSE: 0.1585 -> 0.4961 | Hurt |
| Use modified pushforward training (always on, eval mode, fixed sample count) | OrbEvo-DM-s8 · MDA | Rollout nRMSE: 0.1778 -> 0.1613 | Improved |
| Use modified pushforward training | OrbEvo-WF-sall · MDA | Rollout nRMSE: 0.1585 -> 0.2008 | Hurt |
Closest papers by meaning
- Gaussian Plane-Wave Neural Operator for Electron Density Estimation (ICML 2024 · similarity 0.86)
- Enhancing the Scalability and Applicability of Kohn-Sham Hamiltonians for Molecular Systems (ICLR 2025 · similarity 0.86)
- ELECTRA: A Cartesian Network for 3D Charge Density Prediction with Floating Orbitals (NeurIPS 2025 · similarity 0.85)
- Neural Pfaffians: Solving Many Many-Electron Schrödinger Equations (NeurIPS 2024 · similarity 0.85)
- Learning Equivariant Non-Local Electron Density Functionals (ICLR 2025 · similarity 0.85)
- High-order Equivariant Flow Matching for Density Functional Theory Hamiltonian Prediction (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.
