2601.07131v1.pdf

{
  "dorsal/arxiv": [
    {
      "date_modified": "2026-02-17T05:53:12.314000Z",
      "hash": "6541c578846124092d11c28dd646c4e520332fac985274da47ac7f549368eb40",
      "id": "866f7e30-20a4-41c1-9de9-c20c12fa4d72",
      "source": {
        "execution_id": "54a103de-30bd-4348-b3b7-555ad7f7b8e8",
        "id": "arXiv Dataset",
        "type": "Model",
        "variant": "snapshot-2026-01-17",
        "version": "0.1.0"
      },
      "user_id": 1000002
    }
  ],
  "file/base": {
    "record": {
      "extension": ".pdf",
      "hash": "6541c578846124092d11c28dd646c4e520332fac985274da47ac7f549368eb40",
      "media_type": "application/pdf",
      "media_type_prefix": "application",
      "name": "2601.07131v1.pdf",
      "size": 836569
    },
    "source": {
      "execution_id": "20157a7b-83db-408c-a498-5901e3cc83c6",
      "id": "file/base",
      "type": "Model",
      "version": "1.0.0"
    }
  },
  "file/pdf": {
    "private": false,
    "record": {
      "keywords": [],
      "page_count": 24,
      "producer": "pikepdf 8.15.1",
      "title": "The Limits of Complexity: Why Feature Engineering Beats Deep Learning in Investor Flow Prediction",
      "version": "1.7"
    },
    "source": {
      "execution_id": "b1871d87-c114-4e38-a8a9-eb7a4a080662",
      "id": "dorsal/pdf",
      "type": "Model",
      "version": "1.1.0"
    }
  }
}