{
  "abstract": "Whether physics foundation models can be usefully deployed on laboratory experiments remains an open question for scientific machine learning (ML). We test this question on the Rayleigh-Taylor instability (RTI), a ubiquitous and demanding fluid instability seen from tabletop flows to supernova explosions, in which small perturbations at a density interface grow into chaotic, multiscale mixing as a lighter fluid accelerates into a heavier one. Standard ML models struggle with RTI, and despite over a century of theoretical, numerical, and experimental work, it carries an unresolved discrepancy between simulation and experiment: the late-time mixing growth rate, $α$, measured in most laboratory experiments ($\\sim$ 0.06-0.07), is roughly three times the value from idealized direct numerical simulations (DNS, $\\sim$ 0.02). The gap's origin remains debated. These properties make RTI a stringent test for a question that matters well beyond RTI: can foundation models trained only on simulations generalise to sparse, messy, and noisy laboratory settings? We finetune Walrus, a foundation model for continuum dynamics, on three or fewer DNS realizations and recover key RTI physics over long rollouts. Applied zero-shot to sliding-barrier laboratory data, the finetuned model leaves the DNS-like regime and enters the observed growth band, having never seen a single experimental sample. These results provide independent, data-driven evidence that initial conditions play a crucial role in the longstanding sim-experiment gap in $α$. The model also generalises zero-shot to stable stratification, a buoyancy regime absent from training, correctly slowing mixing-layer growth. Together, our results show that foundation models can generalise well beyond their training data, predicting laboratory behavior and unseen physical regimes, opening new ways to probe longstanding simulation-experiment gaps.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2606.01470",
  "author_total": 23,
  "authors": [
    {
      "name": "Payel Mukhopadhyay"
    },
    {
      "name": "Stefan S. Nixon"
    },
    {
      "name": "Romain Watteaux"
    },
    {
      "name": "Michael McCabe"
    },
    {
      "name": "Alberto Bietti"
    },
    {
      "name": "Kyunghyun Cho"
    },
    {
      "name": "Cristiana Diaconu"
    },
    {
      "name": "Irina Espejo Morales"
    },
    {
      "name": "David Fouhey"
    },
    {
      "name": "Siavash Golkar"
    },
    {
      "name": "Tom Hehir"
    },
    {
      "name": "Shirley Ho"
    },
    {
      "name": "Jake Kovalic"
    },
    {
      "name": "Geraud Krawezik"
    },
    {
      "name": "Francois Lanusse"
    },
    {
      "name": "Tanya Marwah"
    },
    {
      "name": "Rudy Morel"
    },
    {
      "name": "Mariel Pettee"
    },
    {
      "name": "Helen Qu"
    },
    {
      "name": "Jeff Shen"
    },
    {
      "name": "Hadi Sotoudeh"
    },
    {
      "name": "Stuart B. Dalziel"
    },
    {
      "name": "Miles Cranmer"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/82qakge8hvb5s.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2606.01470",
  "fulltext_url": null,
  "id": "82qakge8hvb5s",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2606.01470",
  "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.AI"
    ],
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "physics.flu-dyn"
    ],
    [
      "openalex",
      "3109|Model Reduction and Neural Networks"
    ]
  ],
  "openalex": "W7163161207",
  "pdf_url": "https://arxiv.org/pdf/2606.01470",
  "pmcid": null,
  "pmid": null,
  "published": "2026-11-04",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/82qakge8hvb5s.ris",
  "sources": [
    {
      "first_seen": "2026-11-05T15:20:55Z",
      "native_id": "2606.01470",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2606.01470"
    }
  ],
  "subjects": [
    {
      "field": "physics",
      "origin": "arxiv:physics.flu-dyn",
      "subfield": "fluid-dynamics"
    }
  ],
  "title": "Emergent Transfer of a Physics Foundation Model from Simulation to Laboratory Turbulence",
  "title_text": "Emergent Transfer of a Physics Foundation Model from Simulation to Laboratory Turbulence",
  "type": "preprint",
  "updated_at": "2026-11-07T18:20:05Z",
  "url": "https://research.draconitic.com/paper/82qakge8hvb5s",
  "venue": "arXiv",
  "versions": []
}