{
  "memory": [
    [
      "No explicit memory",
      "54.52",
      "52.24"
    ],
    [
      "Full memory",
      "68.11",
      "67.72"
    ],
    [
      "GT memory",
      "70.77",
      "71.06"
    ],
    [
      "w/o Frames",
      "64.92",
      "63.72"
    ],
    [
      "w/o Count",
      "68.10",
      "66.62"
    ],
    [
      "w/o Relations",
      "67.91",
      "67.72"
    ],
    [
      "10 s",
      "58.52",
      "57.45"
    ],
    [
      "20 s",
      "60.91",
      "60.12"
    ],
    [
      "30 s",
      "63.20",
      "62.48"
    ],
    [
      "40 s",
      "65.04",
      "64.47"
    ],
    [
      "50 s",
      "66.50",
      "65.95"
    ],
    [
      "Full history",
      "68.11",
      "67.72"
    ],
    [
      "1 s",
      "67.47",
      "66.98"
    ],
    [
      "2 s",
      "67.82",
      "67.34"
    ],
    [
      "3 s",
      "68.13",
      "67.72"
    ],
    [
      "5 s",
      "68.11",
      "67.72"
    ],
    [
      "10 s",
      "66.60",
      "65.83"
    ]
  ],
  "action": [
    [
      "No action",
      "47.45 (+0.00)",
      "35.27 (+0.00)",
      "0.00",
      "0.00",
      "0.00"
    ],
    [
      "SFT only",
      "52.92 (+5.47)",
      "41.87 (+6.60)",
      "14.26",
      "43.41",
      "2.19"
    ],
    [
      "SFT + Bandit (full)",
      "58.28 (+10.83)",
      "48.74 (+13.47)",
      "29.13",
      "71.14",
      "2.02"
    ],
    [
      "SFT + GRPO",
      "57.92 (+10.47)",
      "48.20 (+12.93)",
      "28.34",
      "69.78",
      "2.08"
    ],
    [
      "Pixel visibility",
      "51.39 (+3.94)",
      "39.63 (+4.36)",
      "10.40",
      "35.13",
      "2.31"
    ],
    [
      "+ Full framing",
      "52.55 (+5.10)",
      "43.11 (+7.84)",
      "17.79",
      "50.90",
      "2.20"
    ],
    [
      "+ Two-target evidence",
      "55.81 (+8.36)",
      "45.03 (+9.76)",
      "21.69",
      "58.07",
      "2.14"
    ],
    [
      "+ Historical gain (full)",
      "58.28 (+10.83)",
      "48.74 (+13.47)",
      "29.13",
      "71.14",
      "2.02"
    ],
    [
      "No action (0 steps)",
      "47.45 (+0.00)",
      "35.27 (+0.00)",
      "0.00",
      "0.00",
      "0.00"
    ],
    [
      "At most 1 step",
      "54.61 (+7.16)",
      "43.26 (+7.99)",
      "17.85",
      "50.85",
      "0.94"
    ],
    [
      "At most 2 steps",
      "56.93 (+9.48)",
      "46.73 (+11.46)",
      "25.45",
      "64.84",
      "1.57"
    ],
    [
      "At most 3 steps (full)",
      "58.28 (+10.83)",
      "48.74 (+13.47)",
      "29.13",
      "71.14",
      "2.02"
    ]
  ],
  "other": [
    [
      "GPT-5.2",
      "54.49",
      "50.90",
      "41.30",
      "47.30"
    ],
    [
      "Gemini-3-Pro",
      "55.10",
      "60.90",
      "45.20",
      "50.40"
    ],
    [
      "Qwen3.5-9B",
      "57.90",
      "51.60",
      "36.50",
      "48.20"
    ],
    [
      "VLM-3R-7B",
      "60.90",
      "50.10",
      "27.90",
      "40.50"
    ],
    [
      "Cambrian-S-7B",
      "67.50",
      "49.10",
      "27.10",
      "41.30"
    ],
    [
      "MemAct",
      "65.90",
      "55.30",
      "37.80",
      "52.21"
    ]
  ],
  "sources": [
    [
      "3D-FRONT",
      "135,671",
      "218,983",
      "294,677"
    ],
    [
      "ScanNet",
      "31,304",
      "30,431",
      "0"
    ],
    [
      "ScanNet++",
      "19,160",
      "29,201",
      "0"
    ],
    [
      "ARKitScenes",
      "116,526",
      "124,047",
      "0"
    ]
  ],
  "episodes": [
    [
      "3D-FRONT",
      "4,823",
      "14,534"
    ],
    [
      "ScanNet",
      "1,412",
      "1,511"
    ],
    [
      "ScanNet++",
      "796",
      "1,457"
    ],
    [
      "ARKitScenes",
      "4,681",
      "5,764"
    ],
    [
      "Real-world total",
      "6,889",
      "8,732"
    ]
  ],
  "action_lengths": [
    [
      "1",
      "216,418"
    ],
    [
      "2",
      "46,858"
    ],
    [
      "3",
      "31,401"
    ]
  ],
  "templates": [
    [
      "approached object category",
      "18,733"
    ],
    [
      "became visible category",
      "18,733"
    ],
    [
      "camera motion bucket",
      "18,733"
    ],
    [
      "camera path length m",
      "18,733"
    ],
    [
      "camera rotation deg",
      "18,733"
    ],
    [
      "camera to object distance bucket",
      "18,733"
    ],
    [
      "camera to object distance m",
      "18,733"
    ],
    [
      "closer than before",
      "18,733"
    ],
    [
      "closest category then distance",
      "18,732"
    ],
    [
      "closest category then relation",
      "18,732"
    ],
    [
      "closest neighbor category",
      "18,739"
    ],
    [
      "closest object category",
      "18,732"
    ],
    [
      "disappeared category",
      "18,732"
    ],
    [
      "fixed reference relation",
      "1,618"
    ],
    [
      "multi candidate height",
      "1,316"
    ],
    [
      "object category then later visibility",
      "18,732"
    ],
    [
      "object relative to camera",
      "18,732"
    ],
    [
      "room floor area",
      "18,732"
    ],
    [
      "first appearance order",
      "61,034"
    ],
    [
      "first seen category then timestamp",
      "36,870"
    ],
    [
      "first seen object category",
      "18,435"
    ],
    [
      "first seen timestamp s",
      "18,435"
    ],
    [
      "historical coappearance category",
      "60,843"
    ],
    [
      "historical relative direction",
      "26,743"
    ],
    [
      "last seen timestamp",
      "60,847"
    ],
    [
      "seen count by category",
      "24,399"
    ],
    [
      "single seen object then closest category",
      "47,528"
    ],
    [
      "single seen object then closest distance",
      "47,528"
    ],
    [
      "object size m",
      "128,469"
    ],
    [
      "object to object distance bucket",
      "70,293"
    ],
    [
      "object to object distance m",
      "95,915"
    ]
  ],
  "main": [
    {
      "name": "GPT-6 Astra",
      "current": [
        68.16,
        53.04
      ],
      "memory": [
        64.43,
        68.68
      ],
      "action": [
        59.57,
        49.82
      ],
      "overall": [
        66.74,
        61.15
      ]
    },
    {
      "name": "Gemini-3.8-Flash",
      "current": [
        59.26,
        46.29
      ],
      "memory": [
        56.41,
        59.55
      ],
      "action": [
        52.2,
        43.42
      ],
      "overall": [
        58.17,
        53.14
      ]
    },
    {
      "name": "Uniform options",
      "current": [
        23.22,
        null
      ],
      "memory": [
        22.71,
        null
      ],
      "action": [
        25,
        null
      ],
      "overall": [
        23.11,
        null
      ]
    },
    {
      "name": "Modal answer content",
      "current": [
        32.62,
        29.51
      ],
      "memory": [
        22.36,
        41.64
      ],
      "action": [
        54.76,
        34.04
      ],
      "overall": [
        29.89,
        36.73
      ]
    },
    {
      "name": "Modal option letter",
      "current": [
        26.1,
        null
      ],
      "memory": [
        25.68,
        null
      ],
      "action": [
        27.68,
        null
      ],
      "overall": [
        26,
        null
      ]
    },
    {
      "name": "Qwen3.5-4B",
      "current": [
        36.31,
        22.96
      ],
      "memory": [
        46.63,
        43.98
      ],
      "action": [
        31.5,
        21.44
      ],
      "overall": [
        39.5,
        34.23
      ]
    },
    {
      "name": "Qwen3.5-9B",
      "current": [
        42.82,
        27.11
      ],
      "memory": [
        55.31,
        51.78
      ],
      "action": [
        36.66,
        24.59
      ],
      "overall": [
        46.67,
        40.24
      ]
    },
    {
      "name": "Qwen3.8-27B",
      "current": [
        44.47,
        28.51
      ],
      "memory": [
        57.24,
        53.72
      ],
      "action": [
        39.65,
        27.23
      ],
      "overall": [
        48.44,
        42.1
      ]
    },
    {
      "name": "InternVL3.5-8B",
      "current": [
        37.34,
        23.56
      ],
      "memory": [
        47.73,
        44.98
      ],
      "action": [
        32.43,
        22.17
      ],
      "overall": [
        40.55,
        35.07
      ]
    },
    {
      "name": "InternVL3.5-38B",
      "current": [
        39.74,
        25.03
      ],
      "memory": [
        51.66,
        48.78
      ],
      "action": [
        34.88,
        24.08
      ],
      "overall": [
        43.45,
        37.86
      ]
    },
    {
      "name": "VLM-3R",
      "current": [
        43.52,
        27.71
      ],
      "memory": [
        56.63,
        52.69
      ],
      "action": [
        36.78,
        24.96
      ],
      "overall": [
        47.55,
        40.98
      ]
    },
    {
      "name": "Cambrian-S",
      "current": [
        42.54,
        26.56
      ],
      "memory": [
        54.2,
        51.38
      ],
      "action": [
        35.71,
        24.3
      ],
      "overall": [
        46.11,
        39.8
      ]
    },
    {
      "name": "Cambrian-S-FT",
      "current": [
        66.19,
        41.54
      ],
      "memory": [
        63.46,
        61.85
      ],
      "action": [
        57.76,
        39.87
      ],
      "overall": [
        65.1,
        52.4
      ]
    },
    {
      "name": "StreamingVLM",
      "current": [
        63.24,
        38.06
      ],
      "memory": [
        58.85,
        58.54
      ],
      "action": [
        46.03,
        33.66
      ],
      "overall": [
        61.39,
        48.65
      ]
    },
    {
      "name": "Stream3D-VLM",
      "current": [
        60.42,
        36.78
      ],
      "memory": [
        55.88,
        56.46
      ],
      "action": [
        43.76,
        32.32
      ],
      "overall": [
        58.54,
        46.92
      ]
    },
    {
      "name": "MemAct-4B",
      "current": [
        62.01,
        37.5
      ],
      "memory": [
        64.13,
        63.8
      ],
      "action": [
        51.54,
        37.68
      ],
      "overall": [
        62.43,
        51.87
      ]
    },
    {
      "name": "MemAct-9B",
      "current": [
        67.22,
        52.28
      ],
      "memory": [
        68.11,
        67.72
      ],
      "action": [
        58.28,
        48.74
      ],
      "overall": [
        67.27,
        60.23
      ]
    }
  ],
  "timing": {
    "GPT-6 Astra": {
      "timing": "81.3 / 54.2 /  49.6",
      "latency": "--"
    },
    "Gemini-3.8-Flash": {
      "timing": "67.1 / 38.9 / 36.0",
      "latency": "--"
    },
    "Uniform options": {
      "timing": "--",
      "latency": "--"
    },
    "Modal answer content": {
      "timing": "--",
      "latency": "--"
    },
    "Modal option letter": {
      "timing": "--",
      "latency": "--"
    },
    "Qwen3.5-4B": {
      "timing": "30.9 / 12.2 / 10.5",
      "latency": "1.558"
    },
    "Qwen3.5-9B": {
      "timing": "42.3 / 19.8 / 16.7",
      "latency": "2.475"
    },
    "Qwen3.8-27B": {
      "timing": "46.4 / 22.4 / 19.5",
      "latency": "5.423"
    },
    "InternVL3.5-8B": {
      "timing": "33.9 / 13.7 / 11.8",
      "latency": "2.495"
    },
    "InternVL3.5-38B": {
      "timing": "38.8 / 16.8 / 14.6",
      "latency": "5.446"
    },
    "VLM-3R": {
      "timing": "44.0 / 21.0 / 17.5",
      "latency": "2.622"
    },
    "Cambrian-S": {
      "timing": "43.4 / 20.1 / 16.8",
      "latency": "2.446"
    },
    "Cambrian-S-FT": {
      "timing": "73.3 / 47.6 / 38.5",
      "latency": "2.523"
    },
    "StreamingVLM": {
      "timing": "63.6 / 39.1 / 31.3",
      "latency": "0.377"
    },
    "Stream3D-VLM": {
      "timing": "61.0 / 35.7 / 29.2",
      "latency": "0.410"
    },
    "MemAct-4B": {
      "timing": "84.9 / 53.1 / 44.0",
      "latency": "0.383"
    },
    "MemAct-9B": {
      "timing": "85.3 /  57.4 / 50.9",
      "latency": "0.420"
    }
  }
}