mlsanityData to 14 Oct 2026

Coupled-cluster molecular properties across the main group that extrapolate beyond training size

arXiv 2608.18346 ↗First posted 18 Aug 2026physics.chem-ph cond-mat.mtrl-sci cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Coupled-cluster theory defines the accuracy standard for molecular electronic-structure properties but scales too steeply for routine application, whereas density-functional theory is affordable yet systematically biased. We resolve this trade-off with a single equivariant network, HARP (Hamiltonian Read-out for Properties), that predicts an effective one-electron Hamiltonian from one inexpensive B3LYP/def2-SVP calculation and derives a broad suite of properties from it (energy, optical gap, dipole, quadrupole, polarizability, Mulliken atomic charges, and Mayer bond orders) at coupled-cluster accuracy across nine main-group elements, including the under-served phosphorus, sulfur, and chlorine chemistries. The model is trained on a new in-house dataset of multi-property labels computed at the CCSD(T) level for all nine elements. On a held-out test set, it reduces the error of every property by a factor of 3.8 to 270 relative to semi-local, hybrid, and double-hybrid DFT (referenced to composite CCSD(T)/cc-pVTZ), while adding only ~0.1 s wall time per molecule, delivering coupled-cluster-quality predictions at the cost of a single DFT calculation. Critically, deriving every property from a predicted Hamiltonian rather than pooling per-atom features builds the correct size-scaling into the model architecture: on pi-conjugated oligothiophenes it matches finite-field CCSD polarizability to ~1% and the EOM-CCSD optical gap to ~3% at the largest sizes where those references remain affordable (44 and 37 atoms, where a single CCSD field point already costs ~500x the model's entire inference) and extrapolates the corrected trends to 58-atom chains, a regime where pooling-based architectures fail by construction. Accurate extrapolation is therefore set by the model's inductive bias rather than by the training data.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 11 of its 37 references that are in the map. First posted on arXiv 18 Aug 2026.

Builds on

11 of 37 references.

References · 11

  1. 2021Equivariant message passing for the prediction of tensorial properties and molecular spectracited by 441
  2. 2023EquiformerV2: Improved Equivariant Transformer for Scaling to Higher-Degree Representationscited by 204
  3. 2022e3nn: Euclidean Neural Networkscited by 174
  4. 2023Reducing SO(3) Convolutions to SO(2) for Efficient Equivariant GNNscited by 117
  5. 2021SpookyNet: Learning Force Fields with Electronic Degrees of Freedom and Nonlocal Effectscited by 104
  6. 2019Unifying machine learning and quantum chemistry -- a deep neural network for molecular wavefunctionscited by 73
  7. 2021SE(3)-equivariant prediction of molecular wavefunctions and electronic densitiescited by 67
  8. 2023Efficient and Equivariant Graph Networks for Predicting Quantum Hamiltoniancited by 39
  9. 2020Predicting molecular dipole moments by combining atomic partial charges and atomic dipolescited by 18
  10. 2025Accurate and scalable exchange-correlation with deep learningcited by 8
  11. 2025Learning from the electronic structure of molecules across the periodic tablecited by 7

Built on by

Not cited yet.