{
  "abstract": "This study systematically investigates two multi-fidelity strategies used to train machine-learned force fields (MLFFs)—pre-training/fine-tuning and multi-headed training—and elucidates the mechanisms underpinning their success. For pre-training and fine-tuning, we uncover a log–log linear relationship between pre-trained and fine-tuned accuracies that holds across model architectures, model sizes, and quantum-chemical methods. The success of this approach hinges on the quantity and quality of available pre-training data, and, critically, the inclusion of force labels. We demonstrate that pre-trained representations are inherently method-specific, requiring adaptation of the model backbone during fine-tuning. In contrast, multi-headed models learn method-independent backbone representations, where again the heads’ accuracies are log–log linearly related. Relative to pre-training and fine-tuning, these shared representations marginally reduce model performance in most cases. However, this trade-off is offset by practical advantages: multi-headed training extends naturally to multiple labelling methods and enables partial replacement of expensive labels with cheaper alternatives, paving the way towards cost-efficient universal MLFFs.",
  "abstract_shown": true,
  "affiliations": [
    "University of Oxford",
    "Cambridge CB1 2FB"
  ],
  "article_type": null,
  "arxiv": null,
  "author_total": 5,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "John L. A. Gardner",
      "orcid": "0009-0006-7377-7146"
    },
    {
      "affs": [
        1
      ],
      "name": "Hannes Schulz",
      "orcid": "0000-0001-6408-9794"
    },
    {
      "affs": [
        1
      ],
      "name": "Jean Hélie",
      "orcid": "0000-0003-3413-6865"
    },
    {
      "affs": [
        1
      ],
      "name": "Lixin Sun",
      "orcid": "0000-0002-7971-5222"
    },
    {
      "affs": [
        1
      ],
      "name": "Gregor N. C. Simm",
      "orcid": "0000-0001-6815-352X"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/cygqyy0kscpd7.bib",
  "cited_by": 8,
  "doi": "10.1088/2632-2153/ae8e2e",
  "fulltext_url": null,
  "id": "cygqyy0kscpd7",
  "issns": [
    "2632-2153"
  ],
  "landing_url": "https://doi.org/10.1088/2632-2153/ae8e2e",
  "language": "en",
  "licence": "CC-BY",
  "licence_label": "CC BY",
  "licence_source": "journal",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "openalex",
      "2505|Machine Learning in Materials Science"
    ]
  ],
  "openalex": "W4415333337",
  "pdf_url": "https://iopscience.iop.org/article/10.1088/2632-2153/ae8e2e/pdf",
  "pmcid": null,
  "pmid": null,
  "published": "2026-07-21",
  "publisher": "IOP Publishing",
  "ris": "https://research.draconitic.com/paper/cygqyy0kscpd7.ris",
  "sources": [
    {
      "first_seen": "2026-11-04T13:21:20Z",
      "native_id": "W4415333337",
      "source": "openalex",
      "url": "https://doi.org/10.1088/2632-2153/ae8e2e"
    }
  ],
  "subjects": [
    {
      "field": "chemistry",
      "origin": "openalex:2505|Machine Learning in Materials Science",
      "subfield": "materials-polymers"
    }
  ],
  "title": "Understanding multi-fidelity training of machine-learned force-fields",
  "title_text": "Understanding multi-fidelity training of machine-learned force-fields",
  "type": "peer-reviewed",
  "updated_at": "2026-11-09T13:21:22Z",
  "url": "https://research.draconitic.com/paper/cygqyy0kscpd7",
  "venue": "Machine Learning Science and Technology",
  "versions": []
}