2601.08629v1.pdf

{
  "dorsal/arxiv": [
    {
      "date_modified": "2026-02-17T05:53:16.143000Z",
      "hash": "2d4059a8c7f4bb3e4d17fc66fde7cffcffb4261b9938a10a6856d9f87f80cdd3",
      "id": "6e7f09bc-0b90-40ab-b70b-88e0b6984e4c",
      "source": {
        "execution_id": "3a7ccfa0-5722-4762-be71-59f37b16ef74",
        "id": "arXiv Dataset",
        "type": "Model",
        "variant": "snapshot-2026-01-17",
        "version": "0.1.0"
      },
      "user_id": 1000002
    }
  ],
  "file/base": {
    "record": {
      "extension": ".pdf",
      "hash": "2d4059a8c7f4bb3e4d17fc66fde7cffcffb4261b9938a10a6856d9f87f80cdd3",
      "media_type": "application/pdf",
      "media_type_prefix": "application",
      "name": "2601.08629v1.pdf",
      "size": 3550801
    },
    "source": {
      "execution_id": "92a5e220-04fb-4902-a720-bb3b2f67aaa6",
      "id": "file/base",
      "type": "Model",
      "version": "1.0.0"
    }
  },
  "file/pdf": {
    "private": false,
    "record": {
      "keywords": [],
      "page_count": 61,
      "producer": "pikepdf 8.15.1",
      "title": "Get away with less: Need of source side data curation to build parallel corpus for low resource Machine Translation",
      "version": "1.7"
    },
    "source": {
      "execution_id": "71e48d25-f1bd-40d9-9342-6d2dc3dafae3",
      "id": "dorsal/pdf",
      "type": "Model",
      "version": "1.1.0"
    }
  }
}