2601.19821v1.pdf

{
  "file/base": {
    "record": {
      "extension": ".pdf",
      "hash": "dc628446fc409082d9cb9ff1976fc6aeb61f9168938167e58e6f8f862af8538a",
      "media_type": "application/pdf",
      "media_type_prefix": "application",
      "name": "2601.19821v1.pdf",
      "size": 8792618
    },
    "source": {
      "execution_id": "1ae22f81-1f26-4a98-95dc-e5ce209c1a3e",
      "id": "file/base",
      "type": "Model",
      "version": "1.0.0"
    }
  },
  "file/pdf": {
    "private": false,
    "record": {
      "keywords": [],
      "page_count": 17,
      "producer": "pikepdf 8.15.1",
      "title": "Query-Guided Spatial-Temporal-Frequency Interaction for Music Audio-Visual Question Answering",
      "version": "1.7"
    },
    "source": {
      "execution_id": "f26bc20f-be06-4020-b57b-9cec6ad14f03",
      "id": "dorsal/pdf",
      "type": "Model",
      "version": "1.1.0"
    }
  }
}