{
  "abstract": "Learned graph simulators provide an efficient alternative to high-fidelity solvers for granular dynamics. However, granular motion depends strongly on inter-granular contact history, which is difficult to preserve when particle contacts form, break, and rearrange. Existing simulators mainly store temporal information in node features or node-level memory. Here we introduce TRACE, a graph-network simulator that stores interaction history directly on contact edges. Each edge maintains a persistent memory updated by attention-based message passing and a gated recurrent unit, while an edge-identity dictionary preserves this memory as the contact graph changes. A physics-structured decoder predicts inter-granular normal and tangential contact forces, enforces the Coulomb friction limit, and applies equal-and-opposite internal forces. The model is trained with single-step pretraining followed by autoregressive rollout fine-tuning. We evaluate TRACE on 2D and 3D granular column-collapse benchmarks. In both cases, TRACE produces stable, physically consistent long-horizon rollouts, closely reproducing the final deposit geometry and the kinetic energy released during collapse. Compared with graph network simulator (GNS) and node-memory graph neural simulator (NMGNS), TRACE reduces long-rollout position error by 31-62% and final-deposit error by 58-89% across the two benchmarks, while using fewer parameters and maintaining near-zero particle interpenetration. TRACE also achieves 12.2$\\times$ and 8.9$\\times$ speedups over the material point method (MPM) reference solver in 2D and 3D, respectively. Our code is available at https://github.com/Data-Driven-Computational-Geotechnics/TRACE.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2609.02991",
  "author_total": 6,
  "authors": [
    {
      "name": "Changjian Zhou"
    },
    {
      "name": "Negin Yousefpour"
    },
    {
      "name": "Jie Qi"
    },
    {
      "name": "Junfeng Fang"
    },
    {
      "name": "Guillermo A. Narsilio"
    },
    {
      "name": "Hans Petter Jostad"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/8mpbdckhxgfpt.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2609.02991",
  "fulltext_url": null,
  "id": "8mpbdckhxgfpt",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2609.02991",
  "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",
      "cs.NA"
    ],
    [
      "arxiv",
      "math.NA"
    ],
    [
      "arxiv",
      "stat.ML"
    ],
    [
      "openalex",
      "3109|Model Reduction and Neural Networks"
    ]
  ],
  "openalex": "W7208782963",
  "pdf_url": "https://arxiv.org/pdf/2609.02991",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-26",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/8mpbdckhxgfpt.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:25:17Z",
      "native_id": "2609.02991",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2609.02991"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:math.NA",
      "subfield": "numerical-analysis"
    },
    {
      "field": "mathematics",
      "origin": "arxiv:stat.ML",
      "subfield": "statistics"
    }
  ],
  "title": "TRACE: A spatiotemporal contact memory graph network simulator for granular dynamics",
  "title_text": "TRACE: A spatiotemporal contact memory graph network simulator for granular dynamics",
  "type": "preprint",
  "updated_at": "2026-10-30T12:25:26Z",
  "url": "https://research.draconitic.com/paper/8mpbdckhxgfpt",
  "venue": "arXiv",
  "versions": []
}