2601.09733v1.pdf

{
  "dorsal/arxiv": [
    {
      "date_modified": "2026-02-17T05:53:23.006000Z",
      "hash": "401361544d7722552b7c4c9d03398040a1b18549b246501f16454ad52e0f52e6",
      "id": "22256fd1-3136-4428-9b9c-83ddd2d83cd0",
      "source": {
        "execution_id": "947dc9ac-ad2c-4be4-bb2c-6102632e8825",
        "id": "arXiv Dataset",
        "type": "Model",
        "variant": "snapshot-2026-01-17",
        "version": "0.1.0"
      },
      "user_id": 1000002
    }
  ],
  "file/base": {
    "record": {
      "extension": ".pdf",
      "hash": "401361544d7722552b7c4c9d03398040a1b18549b246501f16454ad52e0f52e6",
      "media_type": "application/pdf",
      "media_type_prefix": "application",
      "name": "2601.09733v1.pdf",
      "size": 16885870
    },
    "source": {
      "execution_id": "a89f6285-82ae-4839-b4b7-48672e04d4df",
      "id": "file/base",
      "type": "Model",
      "version": "1.0.0"
    }
  },
  "file/pdf": {
    "private": false,
    "record": {
      "keywords": [],
      "page_count": 43,
      "producer": "pikepdf 8.15.1",
      "title": "Closing the Data Loop: Using OpenDataArena to Engineer Superior Training Datasets",
      "version": "1.7"
    },
    "source": {
      "execution_id": "7fadb7ff-27f1-43bd-9181-66d58bfb1c98",
      "id": "dorsal/pdf",
      "type": "Model",
      "version": "1.1.0"
    }
  }
}