2601.09718v1.pdf

{
  "dorsal/arxiv": [
    {
      "date_modified": "2026-02-17T05:53:20.460000Z",
      "hash": "ddaa15d22e02216e743ff52ece26050af3930ba5ad38233d8ac3937e65f8083c",
      "id": "eb664e3a-8a92-4eef-9a4c-d3a9ba2f15d7",
      "source": {
        "execution_id": "0a73e6a0-2cd4-4547-aa00-9b804660debe",
        "id": "arXiv Dataset",
        "type": "Model",
        "variant": "snapshot-2026-01-17",
        "version": "0.1.0"
      },
      "user_id": 1000002
    }
  ],
  "file/base": {
    "record": {
      "extension": ".pdf",
      "hash": "ddaa15d22e02216e743ff52ece26050af3930ba5ad38233d8ac3937e65f8083c",
      "media_type": "application/pdf",
      "media_type_prefix": "application",
      "name": "2601.09718v1.pdf",
      "size": 1708941
    },
    "source": {
      "execution_id": "6b528c18-e36f-42bf-910a-027092195246",
      "id": "file/base",
      "type": "Model",
      "version": "1.0.0"
    }
  },
  "file/pdf": {
    "private": false,
    "record": {
      "keywords": [],
      "page_count": 31,
      "producer": "pikepdf 8.15.1",
      "title": "StatLLaMA: A multi-stage training framework for building a domain-optimized statistical language model",
      "version": "1.7"
    },
    "source": {
      "execution_id": "243456e0-dcc5-4b45-b2a9-1a17fecc82f4",
      "id": "dorsal/pdf",
      "type": "Model",
      "version": "1.1.0"
    }
  }
}