{
  "benchmark": "SAVED-Bench",
  "version": "1.0",
  "model": "Gemini-3.6-Flash",
  "qa_count": 1988,
  "unit": "percent",
  "precision": 1,
  "columns": [
    "Baseline",
    "FloorSAV",
    "Partial GT map",
    "Full GT map"
  ],
  "rows": [
    {
      "category": "Dynamic Relativity",
      "task": "Ego-to-Exo Direction",
      "qa_count": 200,
      "scores": [
        56.0,
        65.0,
        76.5,
        72.0
      ]
    },
    {
      "category": "Dynamic Relativity",
      "task": "Ego-to-Exo Distance",
      "qa_count": 200,
      "scores": [
        34.3,
        34.6,
        47.9,
        45.9
      ]
    },
    {
      "category": "Dynamic Relativity",
      "task": "Exo-to-Ego Direction",
      "qa_count": 200,
      "scores": [
        32.0,
        50.5,
        50.5,
        53.5
      ]
    },
    {
      "category": "Dynamic Relativity",
      "task": "Exo-to-Ego Distance",
      "qa_count": 200,
      "scores": [
        46.4,
        44.7,
        57.5,
        55.0
      ]
    },
    {
      "category": "Dynamic Relativity",
      "task": "Task average",
      "qa_count": 800,
      "scores": [
        42.2,
        48.7,
        58.1,
        56.6
      ]
    },
    {
      "category": "Regional",
      "task": "Location Awareness",
      "qa_count": 200,
      "scores": [
        92.0,
        94.5,
        96.0,
        96.5
      ]
    },
    {
      "category": "Regional",
      "task": "Visit History",
      "qa_count": 392,
      "scores": [
        86.7,
        92.3,
        89.8,
        97.2
      ]
    },
    {
      "category": "Regional",
      "task": "Task average",
      "qa_count": 592,
      "scores": [
        89.4,
        93.4,
        92.9,
        96.8
      ]
    },
    {
      "category": "Path Reasoning",
      "task": "Line-Path Search (static)",
      "qa_count": 196,
      "scores": [
        60.2,
        77.0,
        78.1,
        77.6
      ]
    },
    {
      "category": "Path Reasoning",
      "task": "Line-Path Search (dynamic)",
      "qa_count": 200,
      "scores": [
        73.0,
        78.0,
        87.5,
        91.0
      ]
    },
    {
      "category": "Path Reasoning",
      "task": "Trajectory Distance",
      "qa_count": 200,
      "scores": [
        16.9,
        25.0,
        25.0,
        28.2
      ]
    },
    {
      "category": "Path Reasoning",
      "task": "Task average",
      "qa_count": 596,
      "scores": [
        50.0,
        60.0,
        63.5,
        65.6
      ]
    },
    {
      "category": "Overall",
      "task": "QA-weighted average",
      "qa_count": 1988,
      "scores": [
        58.3,
        65.3,
        69.8,
        71.3
      ]
    }
  ],
  "qa_sha256": "6c014396e5f1499cff9d8d298ef1d3fdf5f239938fb91ea5decd2730095286b4"
}