2601.15124v2.pdf

{
  "file/base": {
    "record": {
      "extension": ".pdf",
      "hash": "f1468880f46784ef0ac7dd49d65a313f0493cd074776f29701ebabb7771a6cbc",
      "media_type": "application/pdf",
      "media_type_prefix": "application",
      "name": "2601.15124v2.pdf",
      "size": 4995357
    },
    "source": {
      "execution_id": "b2b5449d-dfd1-462b-95b6-0ba1ba5866d3",
      "id": "file/base",
      "type": "Model",
      "version": "1.0.0"
    }
  },
  "file/pdf": {
    "private": false,
    "record": {
      "creation_date": "2026-01-27T01:33:48Z",
      "keywords": [
        "Graph Foundation Models",
        "Retrieval-Augmented Generation",
        "Multi-domain Graph Pre-training",
        "Graph Prompt Learning"
      ],
      "modified_date": "2026-01-27T01:33:48Z",
      "page_count": 12,
      "producer": "pikepdf 8.15.1",
      "subject": "-  Mathematics of computing  -\u003e  Graph algorithms.-  Computing methodologies  -\u003e  Neural networks.Learning latent representations.Knowledge representation and reasoning.",
      "title": "RAG-GFM: Overcoming In-Memory Bottlenecks in Graph Foundation Models via Retrieval-Augmented Generation",
      "version": "1.7"
    },
    "source": {
      "execution_id": "96286475-1a0b-4d77-a978-96542bc9ac94",
      "id": "dorsal/pdf",
      "type": "Model",
      "version": "1.1.0"
    }
  }
}