{
  "abstract": "We introduce machine learning models that accurately predict the electron density, the central variable of density functional theory. They are built on Transformer-inspired architectures, which require little computation per parameter and can therefore be enlarged at low cost. The proposed models, SwinUNETR-V2 and its attention-free counterpart NeXtUNETR, achieve state-of-the-art accuracy on QM9, with normalized mean absolute errors in the density averaging 0.075% and 0.046%, respectively. The predicted densities are accurate enough to bypass the self-consistent iteration: a single Kohn-Sham step reproduces the self-consistent electronic structure. For NeXtUNETR, the median errors are 0.036 meV in the total energy, 0.65 meV/angstrom in the forces and 1.8 meV in the HOMO-LUMO gap. The forces not only recover the self-consistent minima in geometry optimization but also successfully drive ab initio molecular dynamics. Fine-tuned on small datasets, both networks also transfer accurately to systems containing elements absent from QM9 or larger than its molecules.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2610.39687",
  "author_total": 1,
  "authors": [
    {
      "name": "Jeheon Woo"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/a5wv9s3z9q8fg.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.39687",
  "fulltext_url": null,
  "id": "a5wv9s3z9q8fg",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.39687",
  "language": null,
  "licence": "LicenseRef-arXiv-nonexclusive-1.0",
  "licence_label": "arXiv licence",
  "licence_source": "article",
  "licence_url": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "physics.chem-ph"
    ],
    [
      "arxiv",
      "physics.comp-ph"
    ]
  ],
  "openalex": null,
  "pdf_url": "https://arxiv.org/pdf/2610.39687",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-29",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/a5wv9s3z9q8fg.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:36:23Z",
      "native_id": "2610.39687",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.39687"
    }
  ],
  "subjects": [
    {
      "field": "chemistry",
      "origin": "arxiv:physics.chem-ph",
      "subfield": "physical"
    },
    {
      "field": "physics",
      "origin": "arxiv:physics.comp-ph",
      "subfield": "computational"
    }
  ],
  "title": "Transformer-inspired architectures make machine-learned electron densities accurate and transferable enough to bypass self-consistency",
  "title_text": "Transformer-inspired architectures make machine-learned electron densities accurate and transferable enough to bypass self-consistency",
  "type": "preprint",
  "updated_at": "2026-11-01T12:20:45Z",
  "url": "https://research.draconitic.com/paper/a5wv9s3z9q8fg",
  "venue": "arXiv",
  "versions": []
}