2601.09039v1.pdf

{
  "dorsal/arxiv": [
    {
      "date_modified": "2026-02-17T05:53:20.412000Z",
      "hash": "74a4f28fdfaa6acf56dd154b440f302cfa3a52cf0451a4f64182ca1cf23fad9f",
      "id": "0c68ca75-c77e-4f41-a55e-62ca94a009bf",
      "source": {
        "execution_id": "6504b1a2-974d-4893-97a9-aca0cc8e29bb",
        "id": "arXiv Dataset",
        "type": "Model",
        "variant": "snapshot-2026-01-17",
        "version": "0.1.0"
      },
      "user_id": 1000002
    }
  ],
  "file/base": {
    "record": {
      "extension": ".pdf",
      "hash": "74a4f28fdfaa6acf56dd154b440f302cfa3a52cf0451a4f64182ca1cf23fad9f",
      "media_type": "application/pdf",
      "media_type_prefix": "application",
      "name": "2601.09039v1.pdf",
      "size": 11147842
    },
    "source": {
      "execution_id": "ecd1d264-d23c-44b1-9e49-23f4cd3a4313",
      "id": "file/base",
      "type": "Model",
      "version": "1.0.0"
    }
  },
  "file/pdf": {
    "private": false,
    "record": {
      "keywords": [],
      "page_count": 13,
      "producer": "pikepdf 8.15.1",
      "title": "An Information-Theoretic Perspective on LLM Tokenizers",
      "version": "1.7"
    },
    "source": {
      "execution_id": "3eaf81ce-21e5-4395-bfed-29be901f6521",
      "id": "dorsal/pdf",
      "type": "Model",
      "version": "1.1.0"
    }
  }
}