{
  "abstract": "Molecular dynamics predicts molecular motion from atomic forces and trajectory integration. Density functional theory (DFT) provides first-principles forces, but its $O(N^3)$ cost limits simulations to thousands of atoms; empirical force fields scale to millions but miss bond breaking and charge transfer. Equivariant message-passing Transformers learn the ab initio potential at linear cost. Stacked layers propagate information beyond a single cutoff, giving four layers a receptive field up to 23Å. At this frontier, scale is limited by communication: layers exchange boundary features across devices, force backpropagation repeats the exchange, and moving atoms change neighbor graphs and tensor shapes. We present Tessera, a distributed runtime that fixes workload shapes between rebuilds. It sorts atoms along a space-filling curve into equal-count shards, ensuring identical atom counts and tensor shapes per rank. Peer sets scale as $K^{0.5}$, not $K$. A conservative neighbor skin preserves topology across updates, enabling each rebuild interval to replay as one CUDA Graph. Fused kernels, three-level neighbor search, and overlapped All-to-All complete it. On 512 GPUs, Tessera sustains 30.7M atoms at 89.7% weak-scaling efficiency (77--90% across three per-GPU loads); per-step time rises 1.11--1.29x as machine and problem grow 16x. Graph reuse cuts full-step time $1.87\\times$. Across an $8\\times$ device increase, neighbor-search cost grows $1.7\\times$, versus $3.0\\times$ for Morton order. In a live $10.24$M-atom salt solution, NVE energy drift is $0.42\\,μ$eV/atom/ps over $10$ ps, and structure matches a serial reference. A 10.06M-atom Norovirus capsid sustains 100 ps of NVT at $\\approx 8$ ps/day on 256 A100 GPUs. Over 200,000 steps, it has no broken bonds or anomalously close pairs and uses three compiled variants, with no graph breaks or eager fallbacks.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2611.06122",
  "author_total": 13,
  "authors": [
    {
      "name": "Allex Kan"
    },
    {
      "name": "Ji Zhang"
    },
    {
      "name": "Hao Yin"
    },
    {
      "name": "Hao Li"
    },
    {
      "name": "Owen Fu"
    },
    {
      "name": "Xunxing Chen"
    },
    {
      "name": "Lin Huang"
    },
    {
      "name": "Zion Wang"
    },
    {
      "name": "Lucas Li"
    },
    {
      "name": "Jiajun Cheng"
    },
    {
      "name": "Arthur Jiang"
    },
    {
      "name": "Junping Zhao"
    },
    {
      "name": "Jia Zhang"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/3xm6fzj3h84me.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2611.06122",
  "fulltext_url": null,
  "id": "3xm6fzj3h84me",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2611.06122",
  "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",
      "cs.AI"
    ],
    [
      "arxiv",
      "cs.DC"
    ],
    [
      "arxiv",
      "physics.bio-ph"
    ]
  ],
  "openalex": null,
  "pdf_url": "https://arxiv.org/pdf/2611.06122",
  "pmcid": null,
  "pmid": null,
  "published": "2026-11-04",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/3xm6fzj3h84me.ris",
  "sources": [
    {
      "first_seen": "2026-11-05T15:20:56Z",
      "native_id": "2611.06122",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2611.06122"
    }
  ],
  "subjects": [
    {
      "field": "physics",
      "origin": "arxiv:physics.bio-ph",
      "subfield": "biological-medical"
    }
  ],
  "title": "Ab Initio Accuracy Beyond the Cutoff: Distributed Message-Passing Equivariant Attention at Tens-of-Millions-of-Atom Scale",
  "title_text": "Ab Initio Accuracy Beyond the Cutoff: Distributed Message-Passing Equivariant Attention at Tens-of-Millions-of-Atom Scale",
  "type": "preprint",
  "updated_at": "2026-11-09T00:20:19Z",
  "url": "https://research.draconitic.com/paper/3xm6fzj3h84me",
  "venue": "arXiv",
  "versions": []
}