2601.07389v1.pdf

{
  "dorsal/arxiv": [
    {
      "date_modified": "2026-02-17T05:53:12.053000Z",
      "hash": "5d267d77fc8ec0e093b5f2cc75890e40f9455a4ee106b6a1eec6289ede32bb52",
      "id": "97eade30-0be5-4f4f-ab7c-a3d6b7e43477",
      "source": {
        "execution_id": "0ee6b3e7-ad67-4780-8287-929c017340d5",
        "id": "arXiv Dataset",
        "type": "Model",
        "variant": "snapshot-2026-01-17",
        "version": "0.1.0"
      },
      "user_id": 1000002
    }
  ],
  "file/base": {
    "record": {
      "extension": ".pdf",
      "hash": "5d267d77fc8ec0e093b5f2cc75890e40f9455a4ee106b6a1eec6289ede32bb52",
      "media_type": "application/pdf",
      "media_type_prefix": "application",
      "name": "2601.07389v1.pdf",
      "size": 475747
    },
    "source": {
      "execution_id": "eea51ed6-04e6-47e8-b55e-59221e190673",
      "id": "file/base",
      "type": "Model",
      "version": "1.0.0"
    }
  },
  "file/pdf": {
    "private": false,
    "record": {
      "keywords": [],
      "page_count": 14,
      "producer": "pikepdf 8.15.1",
      "title": "On the Non-decoupling of Supervised Fine-tuning and Reinforcement Learning in Post-training",
      "version": "1.7"
    },
    "source": {
      "execution_id": "30df5568-b3d4-4ba0-aa1d-dde148d8f5ed",
      "id": "dorsal/pdf",
      "type": "Model",
      "version": "1.1.0"
    }
  }
}