{
  "abstract": "Learning reliable surrogates for geodynamic models requires benchmarks that resolve coupled fields and temporal stability. We present GeoSim, a benchmark of high-resolution two-dimensional mantle-plume simulations generated with I2VIS, comprising 237 trajectories across two parameter regimes, with six coupled fields, physical timestamps, and trajectory-level splits. GeoSim defines short-window prediction and autonomous rollout, and evaluates representative neural operators field by field. Across models, we observe a pronounced scalar--velocity gap: scalar fields are reconstructed accurately while velocity errors remain large, and short-window accuracy does not imply stable rollout. We also evaluate MFGNO (Multi-Field Graph Neural Operator), a joint multi-field graph baseline designed for this benchmark; on GeoSim-Plume1, under a fixed per-field objective, it reduces horizontal and vertical velocity errors by 58.4% and 38.5% relative to an independent-field graph baseline, while an explicit field-difference input shows no measurable benefit at this scale. GeoSim provides a reproducible benchmark for field-wise diagnosis and long-horizon evaluation of geodynamic surrogates.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2611.04517",
  "author_total": 8,
  "authors": [
    {
      "name": "Hailong Liu"
    },
    {
      "name": "Dinghui Yang"
    },
    {
      "name": "Mengxue Liu"
    },
    {
      "name": "Rui Qi"
    },
    {
      "name": "Meng Zhao"
    },
    {
      "name": "Yihang Tang"
    },
    {
      "name": "Jin Wen"
    },
    {
      "name": "Zhilong Fang"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/9r79j3x88z6pr.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2611.04517",
  "fulltext_url": null,
  "id": "9r79j3x88z6pr",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2611.04517",
  "language": null,
  "licence": "CC-BY-NC-ND-4.0",
  "licence_label": "CC BY-NC-ND 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by-nc-nd/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "physics.geo-ph"
    ]
  ],
  "openalex": null,
  "pdf_url": "https://arxiv.org/pdf/2611.04517",
  "pmcid": null,
  "pmid": null,
  "published": "2026-11-03",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/9r79j3x88z6pr.ris",
  "sources": [
    {
      "first_seen": "2026-11-04T15:21:25Z",
      "native_id": "2611.04517",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2611.04517"
    }
  ],
  "subjects": [
    {
      "field": "earth",
      "origin": "arxiv:physics.geo-ph",
      "subfield": "geophysics"
    }
  ],
  "title": "GeoSim: Benchmarking Multi-Field Neural Operators for Geodynamics",
  "title_text": "GeoSim: Benchmarking Multi-Field Neural Operators for Geodynamics",
  "type": "preprint",
  "updated_at": "2026-11-06T18:21:01Z",
  "url": "https://research.draconitic.com/paper/9r79j3x88z6pr",
  "venue": "arXiv",
  "versions": []
}