2601.21669v1.pdf

{
  "file/base": {
    "record": {
      "extension": ".pdf",
      "hash": "921df39a785df32388d38e5c2dc8c5897049466b84597dc7a86c53f0c01d20a6",
      "media_type": "application/pdf",
      "media_type_prefix": "application",
      "name": "2601.21669v1.pdf",
      "size": 2876359
    },
    "source": {
      "execution_id": "4b69fce3-514c-4ca8-9c26-ca9b84e7be44",
      "id": "file/base",
      "type": "Model",
      "version": "1.0.0"
    }
  },
  "file/pdf": {
    "private": false,
    "record": {
      "keywords": [],
      "page_count": 18,
      "producer": "pikepdf 8.15.1",
      "title": "Expected Return Causes Outcome-Level Mode Collapse in Reinforcement Learning and How to Fix It with Inverse Probability Scaling",
      "version": "1.7"
    },
    "source": {
      "execution_id": "f9b73980-57d9-4f9e-95c4-3a954538d76c",
      "id": "dorsal/pdf",
      "type": "Model",
      "version": "1.1.0"
    }
  }
}