{
  "abstract": "Foundation machine-learned interatomic potentials (MLIPs) promise broad transferability, yet their out-of-domain reliability is hard to establish. Here, we use pure and O/H-containing disordered carbon spanning multiple phases, competing hybridisations and porous frameworks to stress-test six popular foundation MLIPs. We first screen all six on density functional theory (DFT) energy and force errors and binding curves, then subject four to melt-quench-anneal molecular dynamics and evaluate the resulting structural observables. Binding curves and O/H-containing trajectories reveal unphysical extrapolation and dynamical failures that low DFT energy and force errors fail to reveal. Of the models tested, MACE-MH-1 is the most consistently robust, with low DFT errors and stable MD. Despite differing architectures, OMAT-based models share structural biases, highlighting the importance of training coverage. All four models generate structures of nanoporous carbon that reproduce the principal experimental and specialised carbon potential signatures. Incorporating O/H leaves short-range order largely unchanged but shifts pore-size distributions towards larger pores. These results emphasise the need for application-specific validation and extend molecular dynamics modelling of nanoporous carbon towards experimentally realistic structures.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2610.26701",
  "author_total": 4,
  "authors": [
    {
      "name": "Joseph Hart"
    },
    {
      "name": "Samuel G. H. Brookes"
    },
    {
      "name": "Fabian L. Thiemann"
    },
    {
      "name": "Christoph Schran"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/4z3avp2ffav37.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.26701",
  "fulltext_url": null,
  "id": "4z3avp2ffav37",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.26701",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cond-mat.mtrl-sci"
    ],
    [
      "arxiv",
      "physics.chem-ph"
    ],
    [
      "arxiv",
      "physics.comp-ph"
    ],
    [
      "openalex",
      "2505|Machine Learning in Materials Science"
    ]
  ],
  "openalex": "W7223051871",
  "pdf_url": "https://arxiv.org/pdf/2610.26701",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-19",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/4z3avp2ffav37.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:30:52Z",
      "native_id": "2610.26701",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.26701"
    }
  ],
  "subjects": [
    {
      "field": "chemistry",
      "origin": "arxiv:physics.chem-ph",
      "subfield": "physical"
    },
    {
      "field": "physics",
      "origin": "arxiv:physics.comp-ph",
      "subfield": "computational"
    },
    {
      "field": "physics",
      "origin": "arxiv:cond-mat.mtrl-sci",
      "subfield": "condensed-matter"
    }
  ],
  "title": "Challenging Atomistic Foundation Models on Disordered Carbons",
  "title_text": "Challenging Atomistic Foundation Models on Disordered Carbons",
  "type": "preprint",
  "updated_at": "2026-10-30T12:33:41Z",
  "url": "https://research.draconitic.com/paper/4z3avp2ffav37",
  "venue": "arXiv",
  "versions": []
}