2601.10129v1.pdf

{
  "dorsal/arxiv": [
    {
      "date_modified": "2026-02-17T05:53:24.098000Z",
      "hash": "50a531bda46c1c2cae519bc044b690126b829f8940a5f4a7d823af4c964b9d68",
      "id": "fbb56814-879b-4531-9a70-f4918a93bd90",
      "source": {
        "execution_id": "3b41e7f2-50dd-4852-a263-6b26cc8786bc",
        "id": "arXiv Dataset",
        "type": "Model",
        "variant": "snapshot-2026-01-17",
        "version": "0.1.0"
      },
      "user_id": 1000002
    }
  ],
  "file/base": {
    "record": {
      "extension": ".pdf",
      "hash": "50a531bda46c1c2cae519bc044b690126b829f8940a5f4a7d823af4c964b9d68",
      "media_type": "application/pdf",
      "media_type_prefix": "application",
      "name": "2601.10129v1.pdf",
      "size": 1561638
    },
    "source": {
      "execution_id": "3042a60f-9bb3-4066-9c14-279555163a8d",
      "id": "file/base",
      "type": "Model",
      "version": "1.0.0"
    }
  },
  "file/pdf": {
    "private": false,
    "record": {
      "keywords": [],
      "page_count": 14,
      "producer": "pikepdf 8.15.1",
      "title": "LaViT: Aligning Latent Visual Thoughts for Multi-modal Reasoning",
      "version": "1.7"
    },
    "source": {
      "execution_id": "01cf5a1e-b19c-47a1-8dde-931c6772ce8d",
      "id": "dorsal/pdf",
      "type": "Model",
      "version": "1.1.0"
    }
  }
}