Model: gemini-3-pro-preview

sample_run00

Select Configuration:

📊 Sample Metrics

🔍 Exploration

last_node_coverage: 1
last_edge_coverage: 0.390
n_exploration_steps: 9
action_counts:
move: 5
rotate: 7
return: 0
observe: 8
term: 1
forced_term: 0
query: 0
action_cost: 8
final_information_gain: 0.460
action_fail_ratio: 0
valid_action_ratio: 1
false_belief_steps: 6
false_belief_f1: 1
false_belief_f1_position: 1
false_belief_f1_facing: 1
false_belief_action_cost: 5
false_belief_action_cost_after_seen: 0

✅ Evaluation

overall:
n_total: 27
total_score: 19.685
avg_accuracy: 0.729
per_task:
DirectionEvaluationTask:
n_total: 3
task_score: 2.500
avg_accuracy: 0.833
PovEvaluationTask:
n_total: 3
task_score: 1
avg_accuracy: 0.333
BackwardPovTextEvaluationTask:
n_total: 3
task_score: 3
avg_accuracy: 1
View2ActionTextEvaluationTask:
n_total: 3
task_score: 2
avg_accuracy: 0.667
AlloMappingEvaluationTask:
n_total: 3
task_score: 2.575
avg_accuracy: 0.858
RotEvaluationTask:
n_total: 3
task_score: 3
avg_accuracy: 1
Location2ViewEvaluationTask:
n_total: 3
task_score: 2
avg_accuracy: 0.667
View2LocationTextEvaluationTask:
n_total: 3
task_score: 2.609
avg_accuracy: 0.870
Action2ViewEvaluationTask:
n_total: 3
task_score: 1
avg_accuracy: 0.333

✅ Evaluation (prompt_cogmap)

overall:
n_total: 27
total_score: 18.263
avg_accuracy: 0.676
per_task:
DirectionEvaluationTask:
n_total: 3
task_score: 2
avg_accuracy: 0.667
PovEvaluationTask:
n_total: 3
task_score: 2.500
avg_accuracy: 0.833
BackwardPovTextEvaluationTask:
n_total: 3
task_score: 3
avg_accuracy: 1
Action2ViewEvaluationTask:
n_total: 3
task_score: 1
avg_accuracy: 0.333
View2ActionTextEvaluationTask:
n_total: 3
task_score: 1
avg_accuracy: 0.333
AlloMappingEvaluationTask:
n_total: 3
task_score: 2.360
avg_accuracy: 0.787
RotEvaluationTask:
n_total: 3
task_score: 2
avg_accuracy: 0.667
Location2ViewEvaluationTask:
n_total: 3
task_score: 2
avg_accuracy: 0.667
View2LocationTextEvaluationTask:
n_total: 3
task_score: 2.403
avg_accuracy: 0.801

✅ Evaluation (use_gt_cogmap)

overall:
n_total: 27
total_score: 25
avg_accuracy: 0.926
per_task:
RotEvaluationTask:
n_total: 3
task_score: 3
avg_accuracy: 1
Location2ViewEvaluationTask:
n_total: 3
task_score: 3
avg_accuracy: 1
View2LocationTextEvaluationTask:
n_total: 3
task_score: 3
avg_accuracy: 1
DirectionEvaluationTask:
n_total: 3
task_score: 3
avg_accuracy: 1
PovEvaluationTask:
n_total: 3
task_score: 3
avg_accuracy: 1
BackwardPovTextEvaluationTask:
n_total: 3
task_score: 3
avg_accuracy: 1
Action2ViewEvaluationTask:
n_total: 3
task_score: 3
avg_accuracy: 1
View2ActionTextEvaluationTask:
n_total: 3
task_score: 1
avg_accuracy: 0.333
AlloMappingEvaluationTask:
n_total: 3
task_score: 3
avg_accuracy: 1

✅ Evaluation (use_model_cogmap)

overall:
n_total: 27
total_score: 18.964
avg_accuracy: 0.702
per_task:
DirectionEvaluationTask:
n_total: 3
task_score: 2
avg_accuracy: 0.667
PovEvaluationTask:
n_total: 3
task_score: 1.500
avg_accuracy: 0.500
BackwardPovTextEvaluationTask:
n_total: 3
task_score: 3
avg_accuracy: 1
Action2ViewEvaluationTask:
n_total: 3
task_score: 1.500
avg_accuracy: 0.500
View2ActionTextEvaluationTask:
n_total: 3
task_score: 2
avg_accuracy: 0.667
AlloMappingEvaluationTask:
n_total: 3
task_score: 2.079
avg_accuracy: 0.693
RotEvaluationTask:
n_total: 3
task_score: 3
avg_accuracy: 1
Location2ViewEvaluationTask:
n_total: 3
task_score: 1.500
avg_accuracy: 0.500
View2LocationTextEvaluationTask:
n_total: 3
task_score: 2.385
avg_accuracy: 0.795

🧠 Cognitive Map

exploration:
error:
local_vs_gt_local_avg:
dir: 0.833
facing: 1
pos: 0.863
overall: 0.899
global_vs_gt_global_avg:
dir: 0.643
facing: 0.988
pos: 0.708
overall: 0.780
agent_vs_gt_agent_avg:
dir: 0.889
facing: 1
pos: 0.791
overall: 0.893
newly_observed_vs_gt_local_avg:
dir: 0.861
facing: 1
pos: 0.854
overall: 0.905
correctness:
last_global_vs_gt_full:
dir: 0.567
facing: 1
pos: 0.628
overall: 0.732
consistency:
local_vs_global_avg:
dir: 0.750
facing: 1
pos: 0.872
overall: 0.874
position_update_avg: 0.650
facing_update_avg: 0.979
position_stability_avg: 0.650
facing_stability_avg: 0.979
fog_probe:
f1_avg: 0.876
precision_avg: 0.875
recall_avg: 0.889
cogmap_fb:
metrics:
changed:
dir: None
pos: 0.835
facing: 1
overall: None
retention:
dir: None
pos: 0.284
facing: 0
overall: None
unchanged:
dir: 0.655
facing: 0.886
pos: 0.708
overall: 0.749
inertia: -0.000
inertia_list: [-0.00018250211208554782, 5.117662725340122e-05]
unchanged_retention:
dir: None
pos: 0.839
facing: 1
overall: None
unchanged_retention_minus_retention:
pos: 0.885
facing: 1
unchanged_exploration:
dir: 0.564
facing: 1
pos: 0.662
overall: 0.742

🧭 False Belief CogMap

inertia: -0.000
changed:
dir: None
pos: 0.835
facing: 1
overall: None
retention:
dir: None
pos: 0.284
facing: 0
overall: None
unchanged:
dir: 0.655
facing: 0.886
pos: 0.708
overall: 0.749
unchanged_retention:
dir: None
pos: 0.839
facing: 1
overall: None
unchanged_retention_minus_retention:
pos: 0.885
facing: 1
unchanged_exploration:
dir: 0.564
facing: 1
pos: 0.662
overall: 0.742

📊 Performance Charts

Cognitive Map (Update)
Global Update Metrics
Cognitive Map (Full)
Global Full Metrics
Cognitive Map (Self-Tracking)
Global Self-Tracking Metrics
FB CogMap (Unchanged)
False Belief CogMap Unchanged per Turn
Fog Probe F1
Fog Probe F1 per Turn
Fog Probe Precision
Fog Probe Precision per Turn
Fog Probe Recall
Fog Probe Recall per Turn
Position Update
Position Update per Turn
Facing Update
Facing Update per Turn
Position Stability
Position Stability per Turn
Facing Stability
Facing Stability per Turn

🌍 Exploration Phase

🔄 Turn 1

💬 Assistant Action
Actions: [Observe()]
📊 Cognitive Map JSONs
🤖 Predicted
{
  "agent": {
    "position": [
      0,
      0
    ],
    "facing": "north"
  },
  "lamp": {
    "position": [
      0,
      3
    ]
  },
  "bike": {
    "position": [
      2,
      2
    ],
    "facing": "west"
  },
  "blue door": {
    "position": [
      2,
      5
    ]
  }
}
🎯 Ground Truth (Observed)
{
  "bike": {
    "position": [
      2,
      3
    ],
    "facing": "west"
  },
  "lamp": {
    "position": [
      0,
      4
    ],
    "facing": "east"
  },
  "blue door": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  },
  "agent": {
    "position": [
      0,
      0
    ],
    "facing": "north"
  }
}
🎯 Ground Truth (Full)
{
  "television": {
    "position": [
      2,
      -1
    ],
    "facing": "east"
  },
  "chair": {
    "position": [
      2,
      7
    ],
    "facing": "east"
  },
  "truck": {
    "position": [
      10,
      1
    ],
    "facing": "west"
  },
  "bike": {
    "position": [
      2,
      3
    ],
    "facing": "west"
  },
  "backpack": {
    "position": [
      0,
      9
    ],
    "facing": "south"
  },
  "pan": {
    "position": [
      8,
      1
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      4,
      -1
    ],
    "facing": "east"
  },
  "lamp": {
    "position": [
      0,
      4
    ],
    "facing": "east"
  },
  "vase": {
    "position": [
      0,
      11
    ],
    "facing": "east"
  },
  "shelf": {
    "position": [
      12,
      -1
    ],
    "facing": "east"
  },
  "rubberduck": {
    "position": [
      4,
      11
    ],
    "facing": "west"
  },
  "laptop": {
    "position": [
      8,
      -1
    ],
    "facing": "east"
  },
  "green door": {
    "position": [
      6,
      3
    ],
    "facing": "east"
  },
  "blue door": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  },
  "agent": {
    "position": [
      0,
      0
    ],
    "facing": "north"
  }
}
📊 Cognitive Map JSONs
🤖 Predicted
{
  "origin": "agent",
  "objects": {
    "lamp": {
      "position": [
        0,
        3
      ]
    },
    "bike": {
      "position": [
        2,
        3
      ],
      "facing": "-x"
    },
    "blue door": {
      "position": [
        3,
        5
      ]
    }
  }
}
🎯 Ground Truth
{
  "lamp": {
    "position": [
      0,
      4
    ],
    "facing": "east"
  },
  "bike": {
    "position": [
      2,
      3
    ],
    "facing": "west"
  },
  "blue door": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  }
}
📊 Cognitive Map JSONs
🤖 Predicted
{
  "origin": "agent",
  "objects": {
    "lamp": {
      "position": [
        0,
        3
      ]
    },
    "bike": {
      "position": [
        2,
        3
      ],
      "facing": "-x"
    },
    "blue door": {
      "position": [
        3,
        5
      ]
    }
  }
}
🎯 Ground Truth
{
  "lamp": {
    "position": [
      0,
      4
    ],
    "facing": "east"
  },
  "bike": {
    "position": [
      2,
      3
    ],
    "facing": "west"
  },
  "blue door": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  }
}
🗺️ Symbolic Fog Map
###############
#......########
#......########
#......########
#......########
#......########
#......########
######+########
#D....C########
#.A.F..+......#
#......#......#
#.....B#......#
#*..E..#......#
#......#......#
########......#
###############
🔍 Fog Probe JSONs
📍 Candidates: A, B, C, D, E, F
🤖 Predicted: B, C, E
🎯 Ground Truth: B, C, E
🧠 Cognitive Map Metrics
Global:
dir: 0.600
facing: 1
pos: 0.818
overall: 0.806
Global (Full):
dir: 0.050
facing: 0.200
pos: 0.218
overall: 0.156
Local:
dir: 0.667
facing: 1
pos: 0.855
overall: 0.841
Local (Newly):
dir: 0.667
facing: 1
pos: 0.855
overall: 0.841
Fog Probe:
precision: 1
recall: 1
overall: 1
📈 Turn Metrics
node_coverage: 0.214
edge_coverage: 0.073
is_action_fail: False
step: 1
action_counts:
move: 0
rotate: 0
return: 0
observe: 1
term: 0
forced_term: 0
query: 0
information_gain: 0.160
reward: -0.100
is_done: False
is_valid_action: True
Current state
State at Turn 1

🔄 Turn 2

💬 Assistant Action
Actions: [JumpTo(blue door), Rotate(180), Observe()]
📊 Cognitive Map JSONs
🤖 Predicted
{
  "agent": {
    "position": [
      3,
      5
    ],
    "facing": "south"
  },
  "lamp": {
    "position": [
      0,
      3
    ]
  },
  "bike": {
    "position": [
      2,
      2
    ],
    "facing": "west"
  },
  "blue door": {
    "position": [
      3,
      5
    ]
  },
  "green door": {
    "position": [
      5,
      3
    ]
  },
  "television": {
    "position": [
      0,
      1
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      2,
      1
    ],
    "facing": "east"
  }
}
🎯 Ground Truth (Observed)
{
  "television": {
    "position": [
      2,
      -1
    ],
    "facing": "east"
  },
  "bike": {
    "position": [
      2,
      3
    ],
    "facing": "west"
  },
  "cap": {
    "position": [
      4,
      -1
    ],
    "facing": "east"
  },
  "lamp": {
    "position": [
      0,
      4
    ],
    "facing": "east"
  },
  "green door": {
    "position": [
      6,
      3
    ],
    "facing": "east"
  },
  "blue door": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  },
  "agent": {
    "position": [
      5,
      5
    ],
    "facing": "south"
  }
}
🎯 Ground Truth (Full)
{
  "television": {
    "position": [
      2,
      -1
    ],
    "facing": "east"
  },
  "chair": {
    "position": [
      2,
      7
    ],
    "facing": "east"
  },
  "truck": {
    "position": [
      10,
      1
    ],
    "facing": "west"
  },
  "bike": {
    "position": [
      2,
      3
    ],
    "facing": "west"
  },
  "backpack": {
    "position": [
      0,
      9
    ],
    "facing": "south"
  },
  "pan": {
    "position": [
      8,
      1
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      4,
      -1
    ],
    "facing": "east"
  },
  "lamp": {
    "position": [
      0,
      4
    ],
    "facing": "east"
  },
  "vase": {
    "position": [
      0,
      11
    ],
    "facing": "east"
  },
  "shelf": {
    "position": [
      12,
      -1
    ],
    "facing": "east"
  },
  "rubberduck": {
    "position": [
      4,
      11
    ],
    "facing": "west"
  },
  "laptop": {
    "position": [
      8,
      -1
    ],
    "facing": "east"
  },
  "green door": {
    "position": [
      6,
      3
    ],
    "facing": "east"
  },
  "blue door": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  },
  "agent": {
    "position": [
      5,
      5
    ],
    "facing": "south"
  }
}
📊 Cognitive Map JSONs
🤖 Predicted
{
  "origin": "agent",
  "objects": {
    "television": {
      "position": [
        3,
        5
      ],
      "facing": "-x"
    },
    "cap": {
      "position": [
        1,
        5
      ],
      "facing": "-x"
    },
    "green door": {
      "position": [
        -2,
        2
      ]
    }
  }
}
🎯 Ground Truth
{
  "green door": {
    "position": [
      -1,
      2
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      1,
      6
    ],
    "facing": "east"
  },
  "television": {
    "position": [
      3,
      6
    ],
    "facing": "east"
  }
}
📊 Cognitive Map JSONs
🤖 Predicted
{
  "origin": "agent",
  "objects": {
    "television": {
      "position": [
        3,
        5
      ],
      "facing": "-x"
    },
    "cap": {
      "position": [
        1,
        5
      ],
      "facing": "-x"
    },
    "green door": {
      "position": [
        -2,
        2
      ]
    }
  }
}
🎯 Ground Truth
{
  "green door": {
    "position": [
      -1,
      2
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      1,
      6
    ],
    "facing": "east"
  },
  "television": {
    "position": [
      3,
      6
    ],
    "facing": "east"
  }
}
🧠 Cognitive Map Metrics
Global:
dir: 0.679
facing: 1
pos: 0.788
overall: 0.822
Global (Full):
dir: 0.158
facing: 0.400
pos: 0.368
overall: 0.309
Local:
dir: 1
facing: 1
pos: 0.886
overall: 0.962
Local (Newly):
dir: 1
facing: 1
pos: 0.886
overall: 0.962
Fog Probe:
(none)
📈 Turn Metrics
node_coverage: 0.429
edge_coverage: 0.146
is_action_fail: False
step: 2
action_counts:
move: 1
rotate: 1
return: 0
observe: 2
term: 0
forced_term: 0
query: 0
information_gain: 0.249
reward: -0.100
is_done: False
is_valid_action: True
Previous state
State before Turn 2
Current state
State at Turn 2

🔄 Turn 3

💬 Assistant Action
Actions: [Rotate(180), Observe()]
📊 Cognitive Map JSONs
🤖 Predicted
{
  "agent": {
    "position": [
      1,
      5
    ],
    "facing": "north"
  },
  "blue door": {
    "position": [
      1,
      5
    ]
  },
  "lamp": {
    "position": [
      0,
      3
    ]
  },
  "bike": {
    "position": [
      2,
      2
    ],
    "facing": "west"
  },
  "green door": {
    "position": [
      4,
      3
    ]
  },
  "cap": {
    "position": [
      0,
      0
    ],
    "facing": "east"
  },
  "television": {
    "position": [
      -2,
      1
    ],
    "facing": "east"
  },
  "vase": {
    "position": [
      -2,
      9
    ]
  },
  "rubberduck": {
    "position": [
      0,
      10
    ],
    "facing": "west"
  }
}
🎯 Ground Truth (Observed)
{
  "television": {
    "position": [
      2,
      -1
    ],
    "facing": "east"
  },
  "bike": {
    "position": [
      2,
      3
    ],
    "facing": "west"
  },
  "cap": {
    "position": [
      4,
      -1
    ],
    "facing": "east"
  },
  "lamp": {
    "position": [
      0,
      4
    ],
    "facing": "east"
  },
  "vase": {
    "position": [
      0,
      11
    ],
    "facing": "east"
  },
  "rubberduck": {
    "position": [
      4,
      11
    ],
    "facing": "west"
  },
  "green door": {
    "position": [
      6,
      3
    ],
    "facing": "east"
  },
  "blue door": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  },
  "agent": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  }
}
🎯 Ground Truth (Full)
{
  "television": {
    "position": [
      2,
      -1
    ],
    "facing": "east"
  },
  "chair": {
    "position": [
      2,
      7
    ],
    "facing": "east"
  },
  "truck": {
    "position": [
      10,
      1
    ],
    "facing": "west"
  },
  "bike": {
    "position": [
      2,
      3
    ],
    "facing": "west"
  },
  "backpack": {
    "position": [
      0,
      9
    ],
    "facing": "south"
  },
  "pan": {
    "position": [
      8,
      1
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      4,
      -1
    ],
    "facing": "east"
  },
  "lamp": {
    "position": [
      0,
      4
    ],
    "facing": "east"
  },
  "vase": {
    "position": [
      0,
      11
    ],
    "facing": "east"
  },
  "shelf": {
    "position": [
      12,
      -1
    ],
    "facing": "east"
  },
  "rubberduck": {
    "position": [
      4,
      11
    ],
    "facing": "west"
  },
  "laptop": {
    "position": [
      8,
      -1
    ],
    "facing": "east"
  },
  "green door": {
    "position": [
      6,
      3
    ],
    "facing": "east"
  },
  "blue door": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  },
  "agent": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  }
}
📊 Cognitive Map JSONs
🤖 Predicted
{
  "origin": "agent",
  "objects": {
    "vase": {
      "position": [
        -3,
        5
      ]
    },
    "rubberduck": {
      "position": [
        -1,
        6
      ],
      "facing": "-x"
    }
  }
}
🎯 Ground Truth
{
  "vase": {
    "position": [
      -5,
      6
    ],
    "facing": "east"
  },
  "rubberduck": {
    "position": [
      -1,
      6
    ],
    "facing": "west"
  }
}
📊 Cognitive Map JSONs
🤖 Predicted
{
  "origin": "agent",
  "objects": {
    "vase": {
      "position": [
        -3,
        5
      ]
    },
    "rubberduck": {
      "position": [
        -1,
        6
      ],
      "facing": "-x"
    }
  }
}
🎯 Ground Truth
{
  "vase": {
    "position": [
      -5,
      6
    ],
    "facing": "east"
  },
  "rubberduck": {
    "position": [
      -1,
      6
    ],
    "facing": "west"
  }
}
🗺️ Symbolic Fog Map
###############
#......########
#..A...########
#.F....########
#.B....########
#..E.C.########
#...D..########
######*########
#......########
#......+......#
#......#......#
#......#......#
#......#......#
#......#......#
########......#
###############
🔍 Fog Probe JSONs
📍 Candidates: A, B, C, D, E, F
🤖 Predicted: B, D, E
🎯 Ground Truth: B, D, E
🧠 Cognitive Map Metrics
Global:
dir: 0.578
facing: 1
pos: 0.668
overall: 0.749
Global (Full):
dir: 0.217
facing: 0.500
pos: 0.401
overall: 0.372
Local:
dir: 0.667
facing: 1
pos: 0.826
overall: 0.831
Local (Newly):
dir: 0.667
facing: 1
pos: 0.826
overall: 0.831
Fog Probe:
precision: 1
recall: 1
overall: 1
📈 Turn Metrics
node_coverage: 0.571
edge_coverage: 0.195
is_action_fail: False
step: 3
action_counts:
move: 1
rotate: 2
return: 0
observe: 3
term: 0
forced_term: 0
query: 0
information_gain: 0.301
reward: -0.100
is_done: False
is_valid_action: True
Previous state
State before Turn 3
Current state
State at Turn 3

🔄 Turn 4

💬 Assistant Action
Actions: [JumpTo(vase), Rotate(180), Observe()]
📊 Cognitive Map JSONs
🤖 Predicted
{
  "agent": {
    "position": [
      -2,
      9
    ],
    "facing": "south"
  },
  "lamp": {
    "position": [
      0,
      3
    ]
  },
  "bike": {
    "position": [
      2,
      2
    ],
    "facing": "west"
  },
  "blue door": {
    "position": [
      2,
      5
    ]
  },
  "television": {
    "position": [
      -2,
      2
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      -1,
      2
    ],
    "facing": "east"
  },
  "green door": {
    "position": [
      6,
      3
    ]
  },
  "vase": {
    "position": [
      -2,
      9
    ]
  },
  "rubberduck": {
    "position": [
      0,
      9
    ],
    "facing": "west"
  },
  "chair": {
    "position": [
      1,
      6
    ],
    "facing": "east"
  },
  "backpack": {
    "position": [
      -2,
      7
    ],
    "facing": "south"
  }
}
🎯 Ground Truth (Observed)
{
  "television": {
    "position": [
      2,
      -1
    ],
    "facing": "east"
  },
  "chair": {
    "position": [
      2,
      7
    ],
    "facing": "east"
  },
  "bike": {
    "position": [
      2,
      3
    ],
    "facing": "west"
  },
  "backpack": {
    "position": [
      0,
      9
    ],
    "facing": "south"
  },
  "cap": {
    "position": [
      4,
      -1
    ],
    "facing": "east"
  },
  "lamp": {
    "position": [
      0,
      4
    ],
    "facing": "east"
  },
  "vase": {
    "position": [
      0,
      11
    ],
    "facing": "east"
  },
  "rubberduck": {
    "position": [
      4,
      11
    ],
    "facing": "west"
  },
  "green door": {
    "position": [
      6,
      3
    ],
    "facing": "east"
  },
  "blue door": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  },
  "agent": {
    "position": [
      0,
      11
    ],
    "facing": "south"
  }
}
🎯 Ground Truth (Full)
{
  "television": {
    "position": [
      2,
      -1
    ],
    "facing": "east"
  },
  "chair": {
    "position": [
      2,
      7
    ],
    "facing": "east"
  },
  "truck": {
    "position": [
      10,
      1
    ],
    "facing": "west"
  },
  "bike": {
    "position": [
      2,
      3
    ],
    "facing": "west"
  },
  "backpack": {
    "position": [
      0,
      9
    ],
    "facing": "south"
  },
  "pan": {
    "position": [
      8,
      1
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      4,
      -1
    ],
    "facing": "east"
  },
  "lamp": {
    "position": [
      0,
      4
    ],
    "facing": "east"
  },
  "vase": {
    "position": [
      0,
      11
    ],
    "facing": "east"
  },
  "shelf": {
    "position": [
      12,
      -1
    ],
    "facing": "east"
  },
  "rubberduck": {
    "position": [
      4,
      11
    ],
    "facing": "west"
  },
  "laptop": {
    "position": [
      8,
      -1
    ],
    "facing": "east"
  },
  "green door": {
    "position": [
      6,
      3
    ],
    "facing": "east"
  },
  "blue door": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  },
  "agent": {
    "position": [
      0,
      11
    ],
    "facing": "south"
  }
}
📊 Cognitive Map JSONs
🤖 Predicted
{
  "origin": "agent",
  "objects": {
    "chair": {
      "position": [
        -3,
        6
      ],
      "facing": "-x"
    },
    "backpack": {
      "position": [
        0,
        2
      ],
      "facing": "+y"
    },
    "blue door": {
      "position": [
        -2,
        7
      ]
    }
  }
}
🎯 Ground Truth
{
  "backpack": {
    "position": [
      0,
      2
    ],
    "facing": "south"
  },
  "chair": {
    "position": [
      -2,
      4
    ],
    "facing": "east"
  },
  "blue door": {
    "position": [
      -5,
      6
    ],
    "facing": "north"
  }
}
📊 Cognitive Map JSONs
🤖 Predicted
{
  "origin": "agent",
  "objects": {
    "chair": {
      "position": [
        -3,
        6
      ],
      "facing": "-x"
    },
    "backpack": {
      "position": [
        0,
        2
      ],
      "facing": "+y"
    },
    "blue door": {
      "position": [
        -2,
        7
      ]
    }
  }
}
🎯 Ground Truth
{
  "backpack": {
    "position": [
      0,
      2
    ],
    "facing": "south"
  },
  "chair": {
    "position": [
      -2,
      4
    ],
    "facing": "east"
  }
}
🧠 Cognitive Map Metrics
Global:
dir: 0.636
facing: 1
pos: 0.675
overall: 0.770
Global (Full):
dir: 0.350
facing: 0.700
pos: 0.495
overall: 0.515
Local:
dir: 0.500
facing: 1
pos: 0.763
overall: 0.754
Local (Newly):
dir: 1
facing: 1
pos: 0.826
overall: 0.942
Fog Probe:
(none)
📈 Turn Metrics
node_coverage: 0.714
edge_coverage: 0.244
is_action_fail: False
step: 4
action_counts:
move: 2
rotate: 3
return: 0
observe: 4
term: 0
forced_term: 0
query: 0
information_gain: 0.338
reward: -0.100
is_done: False
is_valid_action: True
Previous state
State before Turn 4
Current state
State at Turn 4

🔄 Turn 5

💬 Assistant Action
Actions: [JumpTo(blue door), JumpTo(green door), Rotate(-90), Observe()]
📊 Cognitive Map JSONs
🤖 Predicted
{
  "agent": {
    "position": [
      4,
      3
    ],
    "facing": "east"
  },
  "blue door": {
    "position": [
      2,
      5
    ]
  },
  "green door": {
    "position": [
      4,
      3
    ]
  },
  "lamp": {
    "position": [
      0,
      3
    ]
  },
  "bike": {
    "position": [
      2,
      2
    ],
    "facing": "west"
  },
  "television": {
    "position": [
      -1,
      1
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      0,
      1
    ]
  },
  "vase": {
    "position": [
      -1,
      8
    ]
  },
  "rubberduck": {
    "position": [
      1,
      9
    ],
    "facing": "west"
  },
  "chair": {
    "position": [
      3,
      5
    ],
    "facing": "east"
  },
  "backpack": {
    "position": [
      -1,
      7
    ],
    "facing": "south"
  },
  "truck": {
    "position": [
      7,
      0
    ],
    "facing": "west"
  },
  "pan": {
    "position": [
      6,
      1
    ],
    "facing": "east"
  },
  "shelf": {
    "position": [
      8,
      -1
    ]
  }
}
🎯 Ground Truth (Observed)
{
  "television": {
    "position": [
      2,
      -1
    ],
    "facing": "east"
  },
  "chair": {
    "position": [
      2,
      7
    ],
    "facing": "east"
  },
  "truck": {
    "position": [
      10,
      1
    ],
    "facing": "west"
  },
  "bike": {
    "position": [
      2,
      3
    ],
    "facing": "west"
  },
  "backpack": {
    "position": [
      0,
      9
    ],
    "facing": "south"
  },
  "pan": {
    "position": [
      8,
      1
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      4,
      -1
    ],
    "facing": "east"
  },
  "lamp": {
    "position": [
      0,
      4
    ],
    "facing": "east"
  },
  "vase": {
    "position": [
      0,
      11
    ],
    "facing": "east"
  },
  "shelf": {
    "position": [
      12,
      -1
    ],
    "facing": "east"
  },
  "rubberduck": {
    "position": [
      4,
      11
    ],
    "facing": "west"
  },
  "green door": {
    "position": [
      6,
      3
    ],
    "facing": "east"
  },
  "blue door": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  },
  "agent": {
    "position": [
      6,
      3
    ],
    "facing": "east"
  }
}
🎯 Ground Truth (Full)
{
  "television": {
    "position": [
      2,
      -1
    ],
    "facing": "east"
  },
  "chair": {
    "position": [
      2,
      7
    ],
    "facing": "east"
  },
  "truck": {
    "position": [
      10,
      1
    ],
    "facing": "west"
  },
  "bike": {
    "position": [
      2,
      3
    ],
    "facing": "west"
  },
  "backpack": {
    "position": [
      0,
      9
    ],
    "facing": "south"
  },
  "pan": {
    "position": [
      8,
      1
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      4,
      -1
    ],
    "facing": "east"
  },
  "lamp": {
    "position": [
      0,
      4
    ],
    "facing": "east"
  },
  "vase": {
    "position": [
      0,
      11
    ],
    "facing": "east"
  },
  "shelf": {
    "position": [
      12,
      -1
    ],
    "facing": "east"
  },
  "rubberduck": {
    "position": [
      4,
      11
    ],
    "facing": "west"
  },
  "laptop": {
    "position": [
      8,
      -1
    ],
    "facing": "east"
  },
  "green door": {
    "position": [
      6,
      3
    ],
    "facing": "east"
  },
  "blue door": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  },
  "agent": {
    "position": [
      6,
      3
    ],
    "facing": "east"
  }
}
📊 Cognitive Map JSONs
🤖 Predicted
{
  "origin": "agent",
  "objects": {
    "truck": {
      "position": [
        2,
        4
      ],
      "facing": "-y"
    },
    "pan": {
      "position": [
        2,
        3
      ],
      "facing": "+y"
    },
    "shelf": {
      "position": [
        3,
        4
      ]
    }
  }
}
🎯 Ground Truth
{
  "truck": {
    "position": [
      2,
      4
    ],
    "facing": "west"
  },
  "shelf": {
    "position": [
      4,
      6
    ],
    "facing": "east"
  },
  "pan": {
    "position": [
      2,
      2
    ],
    "facing": "east"
  }
}
📊 Cognitive Map JSONs
🤖 Predicted
{
  "origin": "agent",
  "objects": {
    "truck": {
      "position": [
        2,
        4
      ],
      "facing": "-y"
    },
    "pan": {
      "position": [
        2,
        3
      ],
      "facing": "+y"
    },
    "shelf": {
      "position": [
        3,
        4
      ]
    }
  }
}
🎯 Ground Truth
{
  "truck": {
    "position": [
      2,
      4
    ],
    "facing": "west"
  },
  "shelf": {
    "position": [
      4,
      6
    ],
    "facing": "east"
  },
  "pan": {
    "position": [
      2,
      2
    ],
    "facing": "east"
  }
}
🗺️ Symbolic Fog Map
###############
#......########
#......########
#......########
#......########
#......########
#......########
######+########
#......########
#......*....B.#
#......#......#
#......#A.....#
#......#......#
#......#ED.C..#
########.....F#
###############
🔍 Fog Probe JSONs
📍 Candidates: A, B, C, D, E, F
🤖 Predicted: A, D, E
🎯 Ground Truth: A, D, E
🧠 Cognitive Map Metrics
Global:
dir: 0.714
facing: 0.889
pos: 0.707
overall: 0.770
Global (Full):
dir: 0.625
facing: 0.800
pos: 0.660
overall: 0.695
Local:
dir: 0.833
facing: 1
pos: 0.843
overall: 0.892
Local (Newly):
dir: 0.833
facing: 1
pos: 0.843
overall: 0.892
Fog Probe:
precision: 1
recall: 1
overall: 1
📈 Turn Metrics
node_coverage: 0.929
edge_coverage: 0.317
is_action_fail: False
step: 5
action_counts:
move: 4
rotate: 4
return: 0
observe: 5
term: 0
forced_term: 0
query: 0
information_gain: 0.416
reward: -0.100
is_done: False
is_valid_action: True
Previous state
State before Turn 5
Current state
State at Turn 5

🔄 Turn 6

💬 Assistant Action
Actions: [JumpTo(pan), Rotate(-90), Observe()]
📊 Cognitive Map JSONs
🤖 Predicted
{
  "agent": {
    "position": [
      8,
      -1
    ],
    "facing": "north"
  },
  "blue door": {
    "position": [
      3,
      4
    ]
  },
  "green door": {
    "position": [
      6,
      1
    ]
  },
  "lamp": {
    "position": [
      0,
      3
    ]
  },
  "bike": {
    "position": [
      2,
      2
    ],
    "facing": "west"
  },
  "television": {
    "position": [
      1,
      1
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      0,
      1
    ],
    "facing": "east"
  },
  "vase": {
    "position": [
      0,
      8
    ]
  },
  "rubberduck": {
    "position": [
      1,
      9
    ],
    "facing": "west"
  },
  "chair": {
    "position": [
      4,
      7
    ],
    "facing": "east"
  },
  "backpack": {
    "position": [
      0,
      7
    ],
    "facing": "south"
  },
  "pan": {
    "position": [
      8,
      -1
    ],
    "facing": "east"
  },
  "truck": {
    "position": [
      8,
      0
    ],
    "facing": "west"
  },
  "shelf": {
    "position": [
      9,
      -1
    ]
  }
}
🎯 Ground Truth (Observed)
{
  "television": {
    "position": [
      2,
      -1
    ],
    "facing": "east"
  },
  "chair": {
    "position": [
      2,
      7
    ],
    "facing": "east"
  },
  "truck": {
    "position": [
      10,
      1
    ],
    "facing": "west"
  },
  "bike": {
    "position": [
      2,
      3
    ],
    "facing": "west"
  },
  "backpack": {
    "position": [
      0,
      9
    ],
    "facing": "south"
  },
  "pan": {
    "position": [
      8,
      1
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      4,
      -1
    ],
    "facing": "east"
  },
  "lamp": {
    "position": [
      0,
      4
    ],
    "facing": "east"
  },
  "vase": {
    "position": [
      0,
      11
    ],
    "facing": "east"
  },
  "shelf": {
    "position": [
      12,
      -1
    ],
    "facing": "east"
  },
  "rubberduck": {
    "position": [
      4,
      11
    ],
    "facing": "west"
  },
  "green door": {
    "position": [
      6,
      3
    ],
    "facing": "east"
  },
  "blue door": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  },
  "agent": {
    "position": [
      8,
      1
    ],
    "facing": "north"
  }
}
🎯 Ground Truth (Full)
{
  "television": {
    "position": [
      2,
      -1
    ],
    "facing": "east"
  },
  "chair": {
    "position": [
      2,
      7
    ],
    "facing": "east"
  },
  "truck": {
    "position": [
      10,
      1
    ],
    "facing": "west"
  },
  "bike": {
    "position": [
      2,
      3
    ],
    "facing": "west"
  },
  "backpack": {
    "position": [
      0,
      9
    ],
    "facing": "south"
  },
  "pan": {
    "position": [
      8,
      1
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      4,
      -1
    ],
    "facing": "east"
  },
  "lamp": {
    "position": [
      0,
      4
    ],
    "facing": "east"
  },
  "vase": {
    "position": [
      0,
      11
    ],
    "facing": "east"
  },
  "shelf": {
    "position": [
      12,
      -1
    ],
    "facing": "east"
  },
  "rubberduck": {
    "position": [
      4,
      11
    ],
    "facing": "west"
  },
  "laptop": {
    "position": [
      8,
      -1
    ],
    "facing": "east"
  },
  "green door": {
    "position": [
      6,
      3
    ],
    "facing": "east"
  },
  "blue door": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  },
  "agent": {
    "position": [
      8,
      1
    ],
    "facing": "north"
  }
}
📊 Cognitive Map JSONs
🤖 Predicted
{
  "origin": "agent",
  "objects": {
    "green door": {
      "position": [
        -2,
        2
      ],
      "facing": "+x"
    }
  }
}
🎯 Ground Truth
{
  "green door": {
    "position": [
      -2,
      2
    ],
    "facing": "east"
  }
}
📊 Cognitive Map JSONs
🤖 Predicted
{
  "origin": "agent",
  "objects": {
    "green door": {
      "position": [
        -2,
        2
      ],
      "facing": "+x"
    }
  }
}
🎯 Ground Truth
(no data)
🗺️ Symbolic Fog Map
###############
#......########
#......########
#......########
#......########
#......########
#......########
######+########
#......########
#......+......#
#......#......#
#......#.*..D.#
#......#......#
#......#F...A.#
########.BC.E.#
###############
🔍 Fog Probe JSONs
📍 Candidates: A, B, C, D, E, F
🤖 Predicted: B, C, E, F
🎯 Ground Truth: F, B, C
🧠 Cognitive Map Metrics
Global:
dir: 0.667
facing: 1
pos: 0.738
overall: 0.802
Global (Full):
dir: 0.583
facing: 0.900
pos: 0.689
overall: 0.724
Local:
dir: 1
facing: 1
pos: 1
overall: 1
Local (Newly):
(none)
Fog Probe:
precision: 0.750
recall: 1
overall: 0.857
📈 Turn Metrics
node_coverage: 0.929
edge_coverage: 0.317
is_action_fail: False
step: 6
action_counts:
move: 5
rotate: 5
return: 0
observe: 6
term: 0
forced_term: 0
query: 0
information_gain: 0.423
reward: -0.100
is_done: False
is_valid_action: True
Previous state
State before Turn 6
Current state
State at Turn 6

🔄 Turn 7

💬 Assistant Action
Actions: [Rotate(90), Observe()]
📊 Cognitive Map JSONs
🤖 Predicted
{
  "agent": {
    "position": [
      7,
      1
    ],
    "facing": "east"
  },
  "lamp": {
    "position": [
      0,
      2
    ],
    "facing": null
  },
  "bike": {
    "position": [
      2,
      2
    ],
    "facing": "west"
  },
  "blue door": {
    "position": [
      1,
      6
    ],
    "facing": "south"
  },
  "television": {
    "position": [
      -2,
      2
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      0,
      3
    ],
    "facing": "east"
  },
  "green door": {
    "position": [
      5,
      2
    ],
    "facing": "west"
  },
  "vase": {
    "position": [
      -2,
      9
    ],
    "facing": null
  },
  "rubberduck": {
    "position": [
      0,
      9
    ],
    "facing": "west"
  },
  "chair": {
    "position": [
      1,
      7
    ],
    "facing": "east"
  },
  "backpack": {
    "position": [
      -2,
      8
    ],
    "facing": "south"
  },
  "truck": {
    "position": [
      8,
      1
    ],
    "facing": "west"
  },
  "pan": {
    "position": [
      7,
      1
    ],
    "facing": "east"
  },
  "shelf": {
    "position": [
      8,
      0
    ],
    "facing": null
  }
}
🎯 Ground Truth (Observed)
{
  "television": {
    "position": [
      2,
      -1
    ],
    "facing": "east"
  },
  "chair": {
    "position": [
      2,
      7
    ],
    "facing": "east"
  },
  "truck": {
    "position": [
      10,
      1
    ],
    "facing": "west"
  },
  "bike": {
    "position": [
      2,
      3
    ],
    "facing": "west"
  },
  "backpack": {
    "position": [
      0,
      9
    ],
    "facing": "south"
  },
  "pan": {
    "position": [
      8,
      1
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      4,
      -1
    ],
    "facing": "east"
  },
  "lamp": {
    "position": [
      0,
      4
    ],
    "facing": "east"
  },
  "vase": {
    "position": [
      0,
      11
    ],
    "facing": "east"
  },
  "shelf": {
    "position": [
      12,
      -1
    ],
    "facing": "east"
  },
  "rubberduck": {
    "position": [
      4,
      11
    ],
    "facing": "west"
  },
  "green door": {
    "position": [
      6,
      3
    ],
    "facing": "east"
  },
  "blue door": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  },
  "agent": {
    "position": [
      8,
      1
    ],
    "facing": "east"
  }
}
🎯 Ground Truth (Full)
{
  "television": {
    "position": [
      2,
      -1
    ],
    "facing": "east"
  },
  "chair": {
    "position": [
      2,
      7
    ],
    "facing": "east"
  },
  "truck": {
    "position": [
      10,
      1
    ],
    "facing": "west"
  },
  "bike": {
    "position": [
      2,
      3
    ],
    "facing": "west"
  },
  "backpack": {
    "position": [
      0,
      9
    ],
    "facing": "south"
  },
  "pan": {
    "position": [
      8,
      1
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      4,
      -1
    ],
    "facing": "east"
  },
  "lamp": {
    "position": [
      0,
      4
    ],
    "facing": "east"
  },
  "vase": {
    "position": [
      0,
      11
    ],
    "facing": "east"
  },
  "shelf": {
    "position": [
      12,
      -1
    ],
    "facing": "east"
  },
  "rubberduck": {
    "position": [
      4,
      11
    ],
    "facing": "west"
  },
  "laptop": {
    "position": [
      8,
      -1
    ],
    "facing": "east"
  },
  "green door": {
    "position": [
      6,
      3
    ],
    "facing": "east"
  },
  "blue door": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  },
  "agent": {
    "position": [
      8,
      1
    ],
    "facing": "east"
  }
}
📊 Cognitive Map JSONs
🤖 Predicted
{
  "origin": "agent",
  "objects": {
    "truck": {
      "position": [
        0,
        2
      ],
      "facing": "-y"
    },
    "shelf": {
      "position": [
        4,
        4
      ]
    }
  }
}
🎯 Ground Truth
{
  "truck": {
    "position": [
      0,
      2
    ],
    "facing": "west"
  },
  "shelf": {
    "position": [
      2,
      4
    ],
    "facing": "east"
  }
}
📊 Cognitive Map JSONs
🤖 Predicted
{
  "origin": "agent",
  "objects": {
    "truck": {
      "position": [
        0,
        2
      ],
      "facing": "-y"
    },
    "shelf": {
      "position": [
        4,
        4
      ]
    }
  }
}
🎯 Ground Truth
(no data)
🗺️ Symbolic Fog Map
###############
#......########
#......########
#......########
#......########
#......########
#......########
######+########
#......########
#......+...B..#
#......#.F....#
#......#.*....#
#......#..D...#
#......#.CA...#
########E.....#
###############
🔍 Fog Probe JSONs
📍 Candidates: A, B, C, D, E, F
🤖 Predicted: B, E
🎯 Ground Truth: C, E, A
🧠 Cognitive Map Metrics
Global:
dir: 0.714
facing: 1
pos: 0.684
overall: 0.799
Global (Full):
dir: 0.625
facing: 0.900
pos: 0.638
overall: 0.721
Local:
dir: 1
facing: 1
pos: 0.843
overall: 0.948
Local (Newly):
(none)
Fog Probe:
precision: 0.500
recall: 0.333
overall: 0.400
📈 Turn Metrics
node_coverage: 0.929
edge_coverage: 0.366
is_action_fail: False
step: 7
action_counts:
move: 5
rotate: 6
return: 0
observe: 7
term: 0
forced_term: 0
query: 0
information_gain: 0.429
reward: -0.100
is_done: False
is_valid_action: True
Previous state
State before Turn 7
Current state
State at Turn 7

🔄 Turn 8

💬 Assistant Action
Actions: [Rotate(90), Observe()]
📊 Cognitive Map JSONs
🤖 Predicted
{
  "agent": {
    "position": [
      8,
      -1
    ],
    "facing": "south"
  },
  "blue door": {
    "position": [
      1,
      5
    ],
    "facing": "south"
  },
  "green door": {
    "position": [
      5,
      2
    ],
    "facing": "east"
  },
  "lamp": {
    "position": [
      0,
      3
    ],
    "facing": "south"
  },
  "bike": {
    "position": [
      3,
      3
    ],
    "facing": "west"
  },
  "television": {
    "position": [
      -2,
      2
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      -1,
      2
    ],
    "facing": "east"
  },
  "vase": {
    "position": [
      -2,
      8
    ],
    "facing": "south"
  },
  "rubberduck": {
    "position": [
      -1,
      7
    ],
    "facing": "west"
  },
  "chair": {
    "position": [
      0,
      7
    ],
    "facing": "east"
  },
  "backpack": {
    "position": [
      -2,
      7
    ],
    "facing": "south"
  },
  "pan": {
    "position": [
      8,
      -1
    ],
    "facing": "east"
  },
  "truck": {
    "position": [
      10,
      -1
    ],
    "facing": "west"
  },
  "shelf": {
    "position": [
      10,
      -3
    ],
    "facing": "north"
  },
  "laptop": {
    "position": [
      8,
      -2
    ],
    "facing": "east"
  }
}
🎯 Ground Truth (Observed)
{
  "television": {
    "position": [
      2,
      -1
    ],
    "facing": "east"
  },
  "chair": {
    "position": [
      2,
      7
    ],
    "facing": "east"
  },
  "truck": {
    "position": [
      10,
      1
    ],
    "facing": "west"
  },
  "bike": {
    "position": [
      2,
      3
    ],
    "facing": "west"
  },
  "backpack": {
    "position": [
      0,
      9
    ],
    "facing": "south"
  },
  "pan": {
    "position": [
      8,
      1
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      4,
      -1
    ],
    "facing": "east"
  },
  "lamp": {
    "position": [
      0,
      4
    ],
    "facing": "east"
  },
  "vase": {
    "position": [
      0,
      11
    ],
    "facing": "east"
  },
  "shelf": {
    "position": [
      12,
      -1
    ],
    "facing": "east"
  },
  "rubberduck": {
    "position": [
      4,
      11
    ],
    "facing": "west"
  },
  "laptop": {
    "position": [
      8,
      -1
    ],
    "facing": "east"
  },
  "green door": {
    "position": [
      6,
      3
    ],
    "facing": "east"
  },
  "blue door": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  },
  "agent": {
    "position": [
      8,
      1
    ],
    "facing": "south"
  }
}
🎯 Ground Truth (Full)
{
  "television": {
    "position": [
      2,
      -1
    ],
    "facing": "east"
  },
  "chair": {
    "position": [
      2,
      7
    ],
    "facing": "east"
  },
  "truck": {
    "position": [
      10,
      1
    ],
    "facing": "west"
  },
  "bike": {
    "position": [
      2,
      3
    ],
    "facing": "west"
  },
  "backpack": {
    "position": [
      0,
      9
    ],
    "facing": "south"
  },
  "pan": {
    "position": [
      8,
      1
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      4,
      -1
    ],
    "facing": "east"
  },
  "lamp": {
    "position": [
      0,
      4
    ],
    "facing": "east"
  },
  "vase": {
    "position": [
      0,
      11
    ],
    "facing": "east"
  },
  "shelf": {
    "position": [
      12,
      -1
    ],
    "facing": "east"
  },
  "rubberduck": {
    "position": [
      4,
      11
    ],
    "facing": "west"
  },
  "laptop": {
    "position": [
      8,
      -1
    ],
    "facing": "east"
  },
  "green door": {
    "position": [
      6,
      3
    ],
    "facing": "east"
  },
  "blue door": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  },
  "agent": {
    "position": [
      8,
      1
    ],
    "facing": "south"
  }
}
📊 Cognitive Map JSONs
🤖 Predicted
{
  "origin": "agent",
  "objects": {
    "laptop": {
      "position": [
        0,
        1
      ],
      "facing": "-x"
    }
  }
}
🎯 Ground Truth
{
  "laptop": {
    "position": [
      0,
      2
    ],
    "facing": "east"
  }
}
📊 Cognitive Map JSONs
🤖 Predicted
{
  "origin": "agent",
  "objects": {
    "laptop": {
      "position": [
        0,
        1
      ],
      "facing": "-x"
    }
  }
}
🎯 Ground Truth
{
  "laptop": {
    "position": [
      0,
      2
    ],
    "facing": "east"
  }
}
🗺️ Symbolic Fog Map
###############
#......########
#......########
#......########
#......########
#......########
#......########
######+########
#......########
#......+......#
#......#......#
#......#A*....#
#......#B.....#
#......#......#
########......#
###############
🔍 Fog Probe JSONs
📍 Candidates: A, B
🤖 Predicted: A
🎯 Ground Truth: A
🧠 Cognitive Map Metrics
Global:
dir: 0.633
facing: 1
pos: 0.669
overall: 0.767
Global (Full):
dir: 0.633
facing: 1
pos: 0.669
overall: 0.767
Local:
dir: 1
facing: 1
pos: 0.886
overall: 0.962
Local (Newly):
dir: 1
facing: 1
pos: 0.886
overall: 0.962
Fog Probe:
precision: 1
recall: 1
overall: 1
📈 Turn Metrics
node_coverage: 1
edge_coverage: 0.390
is_action_fail: False
step: 8
action_counts:
move: 5
rotate: 7
return: 0
observe: 8
term: 0
forced_term: 0
query: 0
information_gain: 0.460
reward: -0.100
is_done: False
is_valid_action: True
Previous state
State before Turn 8
Current state
State at Turn 8

🔄 Turn 9

💬 Assistant Action
Actions: [Term()]
📊 Cognitive Map JSONs
🤖 Predicted
{
  "agent": {
    "position": [
      7,
      -1
    ],
    "facing": "south"
  },
  "lamp": {
    "position": [
      0,
      3
    ]
  },
  "bike": {
    "position": [
      2,
      2
    ],
    "facing": "west"
  },
  "blue door": {
    "position": [
      2,
      5
    ]
  },
  "green door": {
    "position": [
      5,
      2
    ]
  },
  "television": {
    "position": [
      -2,
      3
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      -4,
      3
    ],
    "facing": "east"
  },
  "vase": {
    "position": [
      -1,
      8
    ]
  },
  "rubberduck": {
    "position": [
      -1,
      7
    ],
    "facing": "west"
  },
  "chair": {
    "position": [
      1,
      6
    ],
    "facing": "east"
  },
  "backpack": {
    "position": [
      -1,
      6
    ],
    "facing": "south"
  },
  "pan": {
    "position": [
      7,
      -1
    ],
    "facing": "east"
  },
  "truck": {
    "position": [
      9,
      -1
    ],
    "facing": "west"
  },
  "shelf": {
    "position": [
      9,
      -3
    ]
  },
  "laptop": {
    "position": [
      7,
      -3
    ],
    "facing": "east"
  }
}
🎯 Ground Truth (Observed)
{
  "television": {
    "position": [
      2,
      -1
    ],
    "facing": "east"
  },
  "chair": {
    "position": [
      2,
      7
    ],
    "facing": "east"
  },
  "truck": {
    "position": [
      10,
      1
    ],
    "facing": "west"
  },
  "bike": {
    "position": [
      2,
      3
    ],
    "facing": "west"
  },
  "backpack": {
    "position": [
      0,
      9
    ],
    "facing": "south"
  },
  "pan": {
    "position": [
      8,
      1
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      4,
      -1
    ],
    "facing": "east"
  },
  "lamp": {
    "position": [
      0,
      4
    ],
    "facing": "east"
  },
  "vase": {
    "position": [
      0,
      11
    ],
    "facing": "east"
  },
  "shelf": {
    "position": [
      12,
      -1
    ],
    "facing": "east"
  },
  "rubberduck": {
    "position": [
      4,
      11
    ],
    "facing": "west"
  },
  "laptop": {
    "position": [
      8,
      -1
    ],
    "facing": "east"
  },
  "green door": {
    "position": [
      6,
      3
    ],
    "facing": "east"
  },
  "blue door": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  },
  "agent": {
    "position": [
      8,
      1
    ],
    "facing": "south"
  }
}
🎯 Ground Truth (Full)
{
  "television": {
    "position": [
      2,
      -1
    ],
    "facing": "east"
  },
  "chair": {
    "position": [
      2,
      7
    ],
    "facing": "east"
  },
  "truck": {
    "position": [
      10,
      1
    ],
    "facing": "west"
  },
  "bike": {
    "position": [
      2,
      3
    ],
    "facing": "west"
  },
  "backpack": {
    "position": [
      0,
      9
    ],
    "facing": "south"
  },
  "pan": {
    "position": [
      8,
      1
    ],
    "facing": "east"
  },
  "cap": {
    "position": [
      4,
      -1
    ],
    "facing": "east"
  },
  "lamp": {
    "position": [
      0,
      4
    ],
    "facing": "east"
  },
  "vase": {
    "position": [
      0,
      11
    ],
    "facing": "east"
  },
  "shelf": {
    "position": [
      12,
      -1
    ],
    "facing": "east"
  },
  "rubberduck": {
    "position": [
      4,
      11
    ],
    "facing": "west"
  },
  "laptop": {
    "position": [
      8,
      -1
    ],
    "facing": "east"
  },
  "green door": {
    "position": [
      6,
      3
    ],
    "facing": "east"
  },
  "blue door": {
    "position": [
      5,
      5
    ],
    "facing": "north"
  },
  "agent": {
    "position": [
      8,
      1
    ],
    "facing": "south"
  }
}
🧠 Cognitive Map Metrics
Global:
dir: 0.567
facing: 1
pos: 0.628
overall: 0.732
Global (Full):
dir: 0.567
facing: 1
pos: 0.628
overall: 0.732
Local:
(none)
Local (Newly):
(none)
Fog Probe:
(none)
📈 Turn Metrics
node_coverage: 1
edge_coverage: 0.390
is_action_fail: False
step: 9
action_counts:
move: 5
rotate: 7
return: 0
observe: 8
term: 1
forced_term: 0
query: 0
information_gain: 0.460
reward: -0.100
is_done: True
is_valid_action: True
Previous state
State before Turn 9
Current state
State at Turn 9

🧭 False Belief Exploration

🔄 FB Turn 10

💬 Assistant Action
Actions: [Observe()]
📈 Turn Metrics
reward: -0.100
is_done: False
is_valid_action: True
🧭 False Belief Cogmap Metrics
Changed: laptop:
dir: None
pos: 0.886
facing: None
overall: None
Retention: laptop:
dir: None
pos: 0.298
facing: None
overall: None
Unchanged (all):
dir: 0.697
facing: 0.714
pos: 0.737
overall: 0.716
🧾 False Belief Object Sets
all_objects: ['backpack', 'bike', 'blue door', 'cap', 'chair', 'green door', 'lamp', 'laptop', 'pan', 'rubberduck', 'shelf', 'television', 'truck', 'vase']
all_changed_objects: ['cap', 'laptop', 'pan', 'vase']
newly_observed_changed (this turn): ['laptop']
unchanged_objects: ['backpack', 'bike', 'blue door', 'chair', 'green door', 'lamp', 'rubberduck', 'shelf', 'television', 'truck']
pred_changed_keys: ['laptop']
pred_unchanged_keys: ['backpack', 'bike', 'blue door', 'cap', 'chair', 'green door', 'lamp', 'laptop', 'pan', 'rubberduck', 'shelf', 'television', 'truck', 'vase']
🧭 False Belief Info
step: 1
correctly_identified_changes: None
f1_overall: None
f1_position: None
f1_facing: None
action_cost: None
action_cost_after_seen: None
ground_truth_changes: [{'name': 'cap', 'ori': True}, {'name': 'pan', 'ori': True}, {'name': 'laptop', 'pos': True}, {'name': 'vase', 'pos': True}]
reported_changes: []
newly_observed_changed_objects: ['laptop']
newly_observed_unchanged_objects: ['lamp', 'bike', 'blue door']
Current state
State at Turn 1

🔄 FB Turn 11

💬 Assistant Action
Actions: [JumpTo(blue door), Rotate(180), Observe()]
📈 Turn Metrics
reward: -0.100
is_done: False
is_valid_action: True
🧭 False Belief Cogmap Metrics
Changed: cap:
dir: None
pos: None
facing: 1
overall: None
Retention: cap:
dir: None
pos: None
facing: 0
overall: None
Unchanged (all):
dir: 0.561
facing: 1
pos: 0.709
overall: 0.756
🧾 False Belief Object Sets
all_objects: ['backpack', 'bike', 'blue door', 'cap', 'chair', 'green door', 'lamp', 'laptop', 'pan', 'rubberduck', 'shelf', 'television', 'truck', 'vase']
all_changed_objects: ['cap', 'laptop', 'pan', 'vase']
newly_observed_changed (this turn): ['cap']
unchanged_objects: ['backpack', 'bike', 'blue door', 'chair', 'green door', 'lamp', 'rubberduck', 'shelf', 'television', 'truck']
pred_changed_keys: ['cap']
pred_unchanged_keys: ['backpack', 'bike', 'blue door', 'cap', 'chair', 'green door', 'lamp', 'laptop', 'pan', 'rubberduck', 'shelf', 'television', 'truck', 'vase']
🧭 False Belief Info
step: 2
correctly_identified_changes: None
f1_overall: None
f1_position: None
f1_facing: None
action_cost: None
action_cost_after_seen: None
ground_truth_changes: [{'name': 'cap', 'ori': True}, {'name': 'pan', 'ori': True}, {'name': 'laptop', 'pos': True}, {'name': 'vase', 'pos': True}]
reported_changes: []
newly_observed_changed_objects: ['cap']
newly_observed_unchanged_objects: ['green door', 'television']
Previous state
State before Turn 2
Current state
State at Turn 2

🔄 FB Turn 12

💬 Assistant Action
Actions: [Rotate(180), Observe()]
📈 Turn Metrics
reward: -0.100
is_done: False
is_valid_action: True
🧭 False Belief Cogmap Metrics
Unchanged (all):
dir: 0.727
facing: 1
pos: 0.749
overall: 0.825
🧾 False Belief Object Sets
all_objects: ['backpack', 'bike', 'blue door', 'cap', 'chair', 'green door', 'lamp', 'laptop', 'pan', 'rubberduck', 'shelf', 'television', 'truck', 'vase']
all_changed_objects: ['cap', 'laptop', 'pan', 'vase']
newly_observed_changed (this turn): []
unchanged_objects: ['backpack', 'bike', 'blue door', 'chair', 'green door', 'lamp', 'rubberduck', 'shelf', 'television', 'truck']
pred_changed_keys: []
pred_unchanged_keys: ['backpack', 'bike', 'blue door', 'cap', 'chair', 'green door', 'lamp', 'laptop', 'pan', 'rubberduck', 'shelf', 'television', 'truck', 'vase']
🧭 False Belief Info
step: 3
correctly_identified_changes: None
f1_overall: None
f1_position: None
f1_facing: None
action_cost: None
action_cost_after_seen: None
ground_truth_changes: [{'name': 'cap', 'ori': True}, {'name': 'pan', 'ori': True}, {'name': 'laptop', 'pos': True}, {'name': 'vase', 'pos': True}]
reported_changes: []
newly_observed_changed_objects: []
newly_observed_unchanged_objects: ['rubberduck']
Previous state
State before Turn 3
Current state
State at Turn 3

🔄 FB Turn 13

💬 Assistant Action
Actions: [JumpTo(rubberduck), Rotate(180), Observe()]
📈 Turn Metrics
reward: -0.100
is_done: False
is_valid_action: True
🧭 False Belief Cogmap Metrics
Unchanged (all):
dir: 0.636
facing: 0.714
pos: 0.640
overall: 0.663
🧾 False Belief Object Sets
all_objects: ['backpack', 'bike', 'blue door', 'cap', 'chair', 'green door', 'lamp', 'laptop', 'pan', 'rubberduck', 'shelf', 'television', 'truck', 'vase']
all_changed_objects: ['cap', 'laptop', 'pan', 'vase']
newly_observed_changed (this turn): []
unchanged_objects: ['backpack', 'bike', 'blue door', 'chair', 'green door', 'lamp', 'rubberduck', 'shelf', 'television', 'truck']
pred_changed_keys: []
pred_unchanged_keys: ['backpack', 'bike', 'blue door', 'cap', 'chair', 'green door', 'lamp', 'laptop', 'pan', 'rubberduck', 'shelf', 'television', 'truck', 'vase']
🧭 False Belief Info
step: 4
correctly_identified_changes: None
f1_overall: None
f1_position: None
f1_facing: None
action_cost: None
action_cost_after_seen: None
ground_truth_changes: [{'name': 'cap', 'ori': True}, {'name': 'pan', 'ori': True}, {'name': 'laptop', 'pos': True}, {'name': 'vase', 'pos': True}]
reported_changes: []
newly_observed_changed_objects: []
newly_observed_unchanged_objects: ['chair']
Previous state
State before Turn 4
Current state
State at Turn 4

🔄 FB Turn 14

💬 Assistant Action
Actions: [JumpTo(blue door), JumpTo(green door), Rotate(-90), Observe()]
📈 Turn Metrics
reward: -0.100
is_done: False
is_valid_action: True
🧭 False Belief Cogmap Metrics
Changed: pan:
dir: None
pos: None
facing: 1
overall: None
Retention: pan:
dir: None
pos: None
facing: 0
overall: None
Changed: vase:
dir: None
pos: 0.785
facing: None
overall: None
Retention: vase:
dir: None
pos: 0.270
facing: None
overall: None
Unchanged (all):
dir: 0.652
facing: 1
pos: 0.706
overall: 0.786
🧾 False Belief Object Sets
all_objects: ['backpack', 'bike', 'blue door', 'cap', 'chair', 'green door', 'lamp', 'laptop', 'pan', 'rubberduck', 'shelf', 'television', 'truck', 'vase']
all_changed_objects: ['cap', 'laptop', 'pan', 'vase']
newly_observed_changed (this turn): ['pan', 'vase']
unchanged_objects: ['backpack', 'bike', 'blue door', 'chair', 'green door', 'lamp', 'rubberduck', 'shelf', 'television', 'truck']
pred_changed_keys: ['pan', 'vase']
pred_unchanged_keys: ['backpack', 'bike', 'blue door', 'cap', 'chair', 'green door', 'lamp', 'laptop', 'pan', 'rubberduck', 'shelf', 'television', 'truck', 'vase']
🧭 False Belief Info
step: 5
correctly_identified_changes: None
f1_overall: None
f1_position: None
f1_facing: None
action_cost: None
action_cost_after_seen: None
ground_truth_changes: [{'name': 'cap', 'ori': True}, {'name': 'pan', 'ori': True}, {'name': 'laptop', 'pos': True}, {'name': 'vase', 'pos': True}]
reported_changes: []
newly_observed_changed_objects: ['pan', 'vase']
newly_observed_unchanged_objects: ['shelf', 'truck']
Previous state
State before Turn 5
Current state
State at Turn 5

🔄 FB Turn 15

💬 Assistant Action
Actions: [Term(changes="laptop: position, vase: position, cap: orientation, pan: orientation")]
📈 Turn Metrics
reward: 1
is_done: True
is_valid_action: True
success: 1
🧭 False Belief Info
step: 6
correctly_identified_changes: 1
f1_overall: 1
f1_position: 1
f1_facing: 1
action_cost: 5
action_cost_after_seen: 0
ground_truth_changes: [{'name': 'cap', 'ori': True}, {'name': 'pan', 'ori': True}, {'name': 'laptop', 'pos': True}, {'name': 'vase', 'pos': True}]
reported_changes: [{'name': 'laptop', 'pos': True}, {'name': 'vase', 'pos': True}, {'name': 'cap', 'ori': True}, {'name': 'pan', 'ori': True}]
newly_observed_changed_objects: []
newly_observed_unchanged_objects: []
Previous state
State before Turn 6
Current state
State at Turn 6

📊 Evaluation (default)

📊 Task: DirectionEvaluationTask

Question 1 (ID: 88ffe47276e19372)

✅ Evaluation Results
id: 88ffe47276e19372
question: You return to your starting position and face north. From a Top-Down map, describe where shelf is relative to truck. Answer format: <cardinal direction>, <distance> Example: north-west, near
action: None
answer: south east, mid distance
task_type: DirectionEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: default
Evaluation state
Q1: DirectionEvaluationTask

Question 2 (ID: 717cb6885cdcac87)

✅ Evaluation Results
id: 717cb6885cdcac87
question: You return to your starting position and face north. From a Top-Down map, describe where rubberduck is relative to shelf. Answer format: <cardinal direction>, <distance> Example: north-west, near
action: None
answer: north west, far
task_type: DirectionEvaluationTask
choices: []
kwargs:
(none)
score: 0.500
evaluation_mode: default
Evaluation state
Q2: DirectionEvaluationTask

Question 3 (ID: c7a5d595404c5cc8)

✅ Evaluation Results
id: c7a5d595404c5cc8
question: You return to your starting position and face north. From a Top-Down map, describe where backpack is relative to pan. Answer format: <cardinal direction>, <distance> Example: north-west, near
action: None
answer: north west, far
task_type: DirectionEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: default
Evaluation state
Q3: DirectionEvaluationTask

📊 Task: PovEvaluationTask

Question 1 (ID: 14625b47c376defb)

✅ Evaluation Results
id: 14625b47c376defb
question: Now you jump to backpack's direction, facing its direction. Describe where chair is relative to you. Answer format: <ego direction>, <distance> Example: front-left, near
action: None
answer: front-left, mid distance
task_type: PovEvaluationTask
choices: []
kwargs:
(none)
score: 0.500
evaluation_mode: default
Evaluation state
Q1: PovEvaluationTask

Question 2 (ID: f49d98e66552e13e)

✅ Evaluation Results
id: f49d98e66552e13e
question: Now you jump to laptop's direction, facing its direction. Describe where shelf is relative to you. Answer format: <ego direction>, <distance> Example: front-left, near
action: None
answer: front, mid distance
task_type: PovEvaluationTask
choices: []
kwargs:
(none)
score: 0
evaluation_mode: default
Evaluation state
Q2: PovEvaluationTask

Question 3 (ID: a211c0863d2dc523)

✅ Evaluation Results
id: a211c0863d2dc523
question: Now you jump to television's direction, facing its direction. Describe where cap is relative to you. Answer format: <ego direction>, <distance> Example: front-left, near
action: None
answer: front, near
task_type: PovEvaluationTask
choices: []
kwargs:
(none)
score: 0.500
evaluation_mode: default
Evaluation state
Q3: PovEvaluationTask

📊 Task: BackwardPovTextEvaluationTask

Question 1 (ID: 6f2581c5107b62fa)

✅ Evaluation Results
id: 6f2581c5107b62fa
question: Now you jump to an object's position, facing its direction. You observe that truck is front-left, mid distance, facing backward; shelf is front, mid distance. Which object are you standing at? Answer format: <object_name> Example: lamp
action: None
answer:
answer: laptop
final_pos: [9, 2]
final_ori: [1, 0]
final_observation: [{'name': 'truck', 'direction': 'front-left', 'distance': 'mid distance', 'orientation': 'facing backward'}, {'name': 'shelf', 'direction': 'front', 'distance': 'mid distance', 'orientation': None}]
object_positions:
television: [3.0, 2.0]
chair: [3.0, 10.0]
truck: [11.0, 4.0]
bike: [3.0, 6.0]
backpack: [1.0, 12.0]
pan: [9.0, 4.0]
cap: [5.0, 2.0]
lamp: [1.0, 7.0]
vase: [1.0, 14.0]
shelf: [13.0, 2.0]
rubberduck: [5.0, 14.0]
laptop: [9.0, 2.0]
green door: [7.0, 6.0]
blue door: [6.0, 8.0]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
room_id: 3
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: BackwardPovTextEvaluationTask
choices: []
kwargs:
(none)
name_match: True
view_match: True
score: 1
evaluation_mode: default
Evaluation state
Q1: BackwardPovTextEvaluationTask

Question 2 (ID: 534ed0980f85c17a)

✅ Evaluation Results
id: 534ed0980f85c17a
question: Now you jump to an object's position, facing its direction. You observe that pan is front, near, facing backward; laptop is front-left, mid distance, facing backward; green door is front-right, slightly far, on front wall. Which object are you standing at? Answer format: <object_name> Example: lamp
action: None
answer:
answer: truck
final_pos: [11, 4]
final_ori: [-1, 0]
final_observation: [{'name': 'pan', 'direction': 'front', 'distance': 'near', 'orientation': 'facing backward'}, {'name': 'laptop', 'direction': 'front-left', 'distance': 'mid distance', 'orientation': 'facing backward'}, {'name': 'green door', 'direction': 'front-right', 'distance': 'slightly far', 'orientation': 'on front wall'}]
object_positions:
television: [3.0, 2.0]
chair: [3.0, 10.0]
truck: [11.0, 4.0]
bike: [3.0, 6.0]
backpack: [1.0, 12.0]
pan: [9.0, 4.0]
cap: [5.0, 2.0]
lamp: [1.0, 7.0]
vase: [1.0, 14.0]
shelf: [13.0, 2.0]
rubberduck: [5.0, 14.0]
laptop: [9.0, 2.0]
green door: [7.0, 6.0]
blue door: [6.0, 8.0]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
room_id: 3
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: BackwardPovTextEvaluationTask
choices: []
kwargs:
(none)
name_match: True
view_match: True
score: 1
evaluation_mode: default
Evaluation state
Q2: BackwardPovTextEvaluationTask

Question 3 (ID: 947e7c0517a89154)

✅ Evaluation Results
id: 947e7c0517a89154
question: Now you jump to an object's position, facing its direction. You observe that truck is front, near, facing backward; shelf is front-right, slightly far. Which object are you standing at? Answer format: <object_name> Example: lamp
action: None
answer:
answer: pan
final_pos: [9, 4]
final_ori: [1, 0]
final_observation: [{'name': 'truck', 'direction': 'front', 'distance': 'near', 'orientation': 'facing backward'}, {'name': 'shelf', 'direction': 'front-right', 'distance': 'slightly far', 'orientation': None}]
object_positions:
television: [3.0, 2.0]
chair: [3.0, 10.0]
truck: [11.0, 4.0]
bike: [3.0, 6.0]
backpack: [1.0, 12.0]
pan: [9.0, 4.0]
cap: [5.0, 2.0]
lamp: [1.0, 7.0]
vase: [1.0, 14.0]
shelf: [13.0, 2.0]
rubberduck: [5.0, 14.0]
laptop: [9.0, 2.0]
green door: [7.0, 6.0]
blue door: [6.0, 8.0]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
room_id: 3
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: BackwardPovTextEvaluationTask
choices: []
kwargs:
(none)
name_match: True
view_match: True
score: 1
evaluation_mode: default
Evaluation state
Q3: BackwardPovTextEvaluationTask

📊 Task: View2ActionTextEvaluationTask

Question 1 (ID: a13a817826736b87)

✅ Evaluation Results
id: a13a817826736b87
question: You return to your starting position and face north. Then you have executed an action sequence and changed to a new location and facing direction. You observe the following: pan is at front-right, slightly far, facing backward; truck is at front-right, mid distance, facing forward; laptop is at front, mid distance, facing backward What action sequence led to this final view? The action sequence must be valid and only contain move actions. Answer format: <sequence of move actions> Example: JumpTo(lamp), Rotate(90)
action: None
answer:
final_pos: [13, 2]
final_ori: [-1, 0]
room_id: 3
init_pos: [1, 3]
init_ori: [0, 1]
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
minimal_plan: [['rotate', 90], ['jumpto', 'green door'], ['jumpto', 'shelf'], ['rotate', 180]]
final_observation: [{'name': 'pan', 'direction': 'front-right', 'distance': 'slightly far', 'orientation': 'facing backward'}, {'name': 'truck', 'direction': 'front-right', 'distance': 'mid distance', 'orientation': 'facing forward'}, {'name': 'laptop', 'direction': 'front', 'distance': 'mid distance', 'orientation': 'facing backward'}]
task_type: View2ActionTextEvaluationTask
choices: []
kwargs:
(none)
pos_match: True
ori_match: True
final_pos: [13.0, 2.0]
final_ori: [-1, 0]
visible_match: True
score: 1
evaluation_mode: default
Evaluation state
Q1: View2ActionTextEvaluationTask

Question 2 (ID: ef2bc20cb1343cb5)

✅ Evaluation Results
id: ef2bc20cb1343cb5
question: You return to your starting position and face north. Then you have executed an action sequence and changed to a new location and facing direction. You observe the following: blue door is at front-left, mid distance, on front wall What action sequence led to this final view? The action sequence must be valid and only contain move actions. Answer format: <sequence of move actions> Example: JumpTo(lamp), Rotate(90)
action: None
answer:
final_pos: [7, 6]
final_ori: [0, 1]
room_id: [1, 3]
init_pos: [1, 3]
init_ori: [0, 1]
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
minimal_plan: [['rotate', 90], ['jumpto', 'initial_pos'], ['rotate', -90]]
final_observation: [{'name': 'blue door', 'direction': 'front-left', 'distance': 'mid distance', 'orientation': 'on front wall'}]
task_type: View2ActionTextEvaluationTask
choices: []
kwargs:
(none)
pos_match: False
ori_match: True
final_pos: [3.0, 6.0]
final_ori: [0, 1]
visible_match: False
score: 0
evaluation_mode: default
Evaluation state
Q2: View2ActionTextEvaluationTask

Question 3 (ID: 3d92d2ea666b2ed5)

✅ Evaluation Results
id: 3d92d2ea666b2ed5
question: You return to your starting position and face north. Then you have executed an action sequence and changed to a new location and facing direction. You observe the following: laptop is at front-left, slightly far, facing left; cap is at front-right, slightly far, facing left; pan is at front-left, mid distance, facing left What action sequence led to this final view? The action sequence must be valid and only contain move actions. Answer format: <sequence of move actions> Example: JumpTo(lamp), Rotate(90)
action: None
answer:
final_pos: [7, 6]
final_ori: [0, -1]
room_id: [1, 3]
init_pos: [1, 3]
init_ori: [0, 1]
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
minimal_plan: [['rotate', 90], ['jumpto', 'initial_pos'], ['rotate', 90]]
final_observation: [{'name': 'laptop', 'direction': 'front-left', 'distance': 'slightly far', 'orientation': 'facing left'}, {'name': 'cap', 'direction': 'front-right', 'distance': 'slightly far', 'orientation': 'facing left'}, {'name': 'pan', 'direction': 'front-left', 'distance': 'mid distance', 'orientation': 'facing left'}]
task_type: View2ActionTextEvaluationTask
choices: []
kwargs:
(none)
pos_match: True
ori_match: True
final_pos: [7.0, 6.0]
final_ori: [0, -1]
visible_match: True
score: 1
evaluation_mode: default
Evaluation state
Q3: View2ActionTextEvaluationTask

📊 Task: AlloMappingEvaluationTask

Question 1 (ID: 47868a3a901e547c)

✅ Evaluation Results
id: 47868a3a901e547c
question: Treat your starting position as the origin (0, 0) while facing north. Report allocentric coordinates using (x right/east, y up/north). Objects: shelf, truck, lamp. Answer format: (x0, y0); (x1, y1); ... in the same order. Example: (1, 0); (-2, 3); (0, -1)
action: None
answer: [[12, -1], [10, 1], [0, 4]]
task_type: AlloMappingEvaluationTask
choices: []
kwargs:
(none)
similarity: 0.859
threshold: 0.900
score: 0.859
evaluation_mode: default
Evaluation state
Q1: AlloMappingEvaluationTask

Question 2 (ID: adbcc906c717830a)

✅ Evaluation Results
id: adbcc906c717830a
question: Treat your starting position as the origin (0, 0) while facing north. Report allocentric coordinates using (x right/east, y up/north). Objects: rubberduck, shelf, pan. Answer format: (x0, y0); (x1, y1); ... in the same order. Example: (1, 0); (-2, 3); (0, -1)
action: None
answer: [[4, 11], [12, -1], [8, 1]]
task_type: AlloMappingEvaluationTask
choices: []
kwargs:
(none)
similarity: 0.866
threshold: 0.900
score: 0.866
evaluation_mode: default
Evaluation state
Q2: AlloMappingEvaluationTask

Question 3 (ID: 9cdfd701876daa0f)

✅ Evaluation Results
id: 9cdfd701876daa0f
question: Treat your starting position as the origin (0, 0) while facing north. Report allocentric coordinates using (x right/east, y up/north). Objects: cap, pan, backpack, shelf. Answer format: (x0, y0); (x1, y1); ... in the same order. Example: (1, 0); (-2, 3); (0, -1)
action: None
answer: [[4, -1], [8, 1], [0, 9], [12, -1]]
task_type: AlloMappingEvaluationTask
choices: []
kwargs:
(none)
similarity: 0.851
threshold: 0.900
score: 0.851
evaluation_mode: default
Evaluation state
Q3: AlloMappingEvaluationTask

📊 Task: RotEvaluationTask

Question 1 (ID: c7b0da8750e06c67)

✅ Evaluation Results
id: c7b0da8750e06c67
question: You return to your starting position and face north. You will perform a full 360-degree rotation by continuously turning counterclockwise in place. Assume all walls are removed (you can see through walls), so every object is visible. Focus on this set of objects: bike, pan, television. List them in the exact order they appear directly ahead while you rotate. If two objects share a bearing, place the nearer one first. Answer format: <object_name1>, <object_name2>, ... Example: mug, sofa, plant
action: None
answer: ['television', 'pan', 'bike']
task_type: RotEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: default
Evaluation state
Q1: RotEvaluationTask

Question 2 (ID: fc92fab04fedd039)

✅ Evaluation Results
id: fc92fab04fedd039
question: You return to your starting position and face north. You will perform a full 360-degree rotation by continuously turning counterclockwise in place. Assume all walls are removed (you can see through walls), so every object is visible. Focus on this set of objects: bike, television, truck. List them in the exact order they appear directly ahead while you rotate. If two objects share a bearing, place the nearer one first. Answer format: <object_name1>, <object_name2>, ... Example: mug, sofa, plant
action: None
answer: ['television', 'truck', 'bike']
task_type: RotEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: default
Evaluation state
Q2: RotEvaluationTask

Question 3 (ID: 83da2dae108d2082)

✅ Evaluation Results
id: 83da2dae108d2082
question: You return to your starting position and face north. You will perform a full 360-degree rotation by continuously turning counterclockwise in place. Assume all walls are removed (you can see through walls), so every object is visible. Focus on this set of objects: chair, television, truck. List them in the exact order they appear directly ahead while you rotate. If two objects share a bearing, place the nearer one first. Answer format: <object_name1>, <object_name2>, ... Example: mug, sofa, plant
action: None
answer: ['television', 'truck', 'chair']
task_type: RotEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: default
Evaluation state
Q3: RotEvaluationTask

📊 Task: RotDualEvaluationTask

Question 1 (ID: cded270d537256ef)

✅ Evaluation Results
id: cded270d537256ef
question: You return to your starting position and face north. You performed a complete 360° rotation in place. Assume all walls are removed (you can see through walls), so every object is visible. During the rotation, these objects appeared directly in front of you in this order: television, pan, bike Based on this sequence, in which direction did you rotate? Answer format: clockwise or counterclockwise Example: clockwise
action: None
answer: counterclockwise
task_type: RotDualEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: default
Evaluation state
Q1: RotDualEvaluationTask

Question 2 (ID: 646028d7ae8a4b2f)

✅ Evaluation Results
id: 646028d7ae8a4b2f
question: You return to your starting position and face north. You performed a complete 360° rotation in place. Assume all walls are removed (you can see through walls), so every object is visible. During the rotation, these objects appeared directly in front of you in this order: television, truck, bike Based on this sequence, in which direction did you rotate? Answer format: clockwise or counterclockwise Example: clockwise
action: None
answer: counterclockwise
task_type: RotDualEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: default
Evaluation state
Q2: RotDualEvaluationTask

Question 3 (ID: 16cddaf912fd8d1e)

✅ Evaluation Results
id: 16cddaf912fd8d1e
question: You return to your starting position and face north. You performed a complete 360° rotation in place. Assume all walls are removed (you can see through walls), so every object is visible. During the rotation, these objects appeared directly in front of you in this order: television, truck, chair Based on this sequence, in which direction did you rotate? Answer format: clockwise or counterclockwise Example: clockwise
action: None
answer: counterclockwise
task_type: RotDualEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: default
Evaluation state
Q3: RotDualEvaluationTask

📊 Task: Location2ViewEvaluationTask

Question 1 (ID: 13d7589043fe1ee0)

✅ Evaluation Results
id: 13d7589043fe1ee0
question: Treat the green door as the new 'origin' (0, 0). You move to (2, -5) and face north. What is the egocentric relation of pan? Answer format: <direction>, <distance> Example: front, near
action: None
answer: front, mid distance
task_type: Action2LocationEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: default
Evaluation state
Q1: Location2ViewEvaluationTask

Question 2 (ID: e2c7ab6a2049c06a)

✅ Evaluation Results
id: e2c7ab6a2049c06a
question: Still treat your initial position as origin (0, 0) You move to (4, 4) and face west. What is the egocentric relation of bike? Answer format: <direction>, <distance> Example: front, near
action: None
answer: front-left, mid distance
task_type: Action2LocationEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: default
Evaluation state
Q2: Location2ViewEvaluationTask

Question 3 (ID: 5a56a83a64962b71)

✅ Evaluation Results
id: 5a56a83a64962b71
question: Still treat your initial position as origin (0, 0) You move to (0, 3) and face east. What is the egocentric relation of blue door? Answer format: <direction>, <distance> Example: front, near
action: None
answer: front-slight-left, slightly far
task_type: Action2LocationEvaluationTask
choices: []
kwargs:
(none)
score: 0
evaluation_mode: default
Evaluation state
Q3: Location2ViewEvaluationTask

📊 Task: View2LocationTextEvaluationTask

Question 1 (ID: 19a422deab9e39fc)

✅ Evaluation Results
id: 19a422deab9e39fc
question: You move to a new location and face north. You observe: pan is at front, mid distance, facing right; truck is at front-right, mid distance, facing left; green door is at front-slight-left, slightly far, on left wall Treat the green door as the new 'origin' (0, 0). What is your new 2D coordinate (x, y)? Answer format: (x, y) Example: (2, -1)
action: None
answer:
coord: [2, -5]
final_pos: [9, 1]
final_ori: [0, 1]
room_id: 3
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
final_observation: [{'name': 'pan', 'direction': 'front', 'distance': 'mid distance', 'orientation': 'facing right'}, {'name': 'truck', 'direction': 'front-right', 'distance': 'mid distance', 'orientation': 'facing left'}, {'name': 'green door', 'direction': 'front-slight-left', 'distance': 'slightly far', 'orientation': 'on left wall'}]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: Location2ActionTextEvaluationTask
choices: []
kwargs:
(none)
similarity: 0.831
raw_score: 0.831
score: 0.831
evaluation_mode: default
Evaluation state
Q1: View2LocationTextEvaluationTask

Question 2 (ID: 197270037eaa3b24)

✅ Evaluation Results
id: 197270037eaa3b24
question: You move to a new location and face west. You observe: bike is at front-left, mid distance, facing forward; lamp is at front, mid distance Still treat your initial position as origin (0, 0) What is your new 2D coordinate (x, y)? Answer format: (x, y) Example: (2, -1)
action: None
answer:
coord: [4, 4]
final_pos: [5, 7]
final_ori: [-1, 0]
room_id: 1
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
final_observation: [{'name': 'bike', 'direction': 'front-left', 'distance': 'mid distance', 'orientation': 'facing forward'}, {'name': 'lamp', 'direction': 'front', 'distance': 'mid distance', 'orientation': None}]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: Location2ActionTextEvaluationTask
choices: []
kwargs:
(none)
similarity: 0.779
raw_score: 0.779
score: 0.779
evaluation_mode: default
Evaluation state
Q2: View2LocationTextEvaluationTask

Question 3 (ID: 7d2a51a882153983)

✅ Evaluation Results
id: 7d2a51a882153983
question: You move to a new location and face east. You observe: blue door is at front-slight-left, slightly far, on left wall; cap is at front-right, slightly far, facing forward; bike is at front, near, facing backward Still treat your initial position as origin (0, 0) What is your new 2D coordinate (x, y)? Answer format: (x, y) Example: (2, -1)
action: None
answer:
coord: [0, 3]
final_pos: [1, 6]
final_ori: [1, 0]
room_id: 1
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
final_observation: [{'name': 'blue door', 'direction': 'front-slight-left', 'distance': 'slightly far', 'orientation': 'on left wall'}, {'name': 'cap', 'direction': 'front-right', 'distance': 'slightly far', 'orientation': 'facing forward'}, {'name': 'bike', 'direction': 'front', 'distance': 'near', 'orientation': 'facing backward'}]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: Location2ActionTextEvaluationTask
choices: []
kwargs:
(none)
similarity: 1
raw_score: 1
score: 1
evaluation_mode: default
Evaluation state
Q3: View2LocationTextEvaluationTask

📊 Task: Action2ViewEvaluationTask

Question 1 (ID: 4f8a7d910b60df45)

✅ Evaluation Results
id: 4f8a7d910b60df45
question: You return to your starting position and face north. You will execute the following action sequence: 1. Jump to the object at front-right, mid distance. 2. Rotate(-90) 3. Jump to the object at front-right, mid distance. 4. Rotate(-180) After executing the actions, what is the ego relation of bike relative to you? Answer format: <ego direction>, <distance> Example: front, near
action: None
answer: front-right, mid distance
task_type: Action2ViewEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: default
Evaluation state
Q1: Action2ViewEvaluationTask

Question 2 (ID: 32029f3c5e3c31ee)

✅ Evaluation Results
id: 32029f3c5e3c31ee
question: You return to your starting position and face north. You will execute the following action sequence: 1. Rotate(90) 2. Among objects which are front-left, slightly far to you, you jump to the 2nd from left also nearest one. 3. Among objects which are front-right, slightly far to you, you jump to the 2nd from left also farthest one. 4. Rotate(-180) After executing the actions, what is the ego relation of truck relative to you? Answer format: <ego direction>, <distance> Example: front, near
action: None
answer: front-right, mid distance
task_type: Action2ViewEvaluationTask
choices: []
kwargs:
(none)
score: 0
evaluation_mode: default
Evaluation state
Q2: Action2ViewEvaluationTask

Question 3 (ID: 79787b772d0b8410)

✅ Evaluation Results
id: 79787b772d0b8410
question: You return to your starting position and face north. You will execute the following action sequence: 1. Jump to the object at front-right, mid distance. 2. Rotate(90) 3. Jump to the object at front-left, mid distance. 4. Rotate(90) After executing the actions, what is the ego relation of cap relative to you? Answer format: <ego direction>, <distance> Example: front, near
action: None
answer: front-slight-right, slightly far
task_type: Action2ViewEvaluationTask
choices: []
kwargs:
(none)
score: 0
evaluation_mode: default
Evaluation state
Q3: Action2ViewEvaluationTask

📊 Evaluation (prompt_cogmap)

📊 Task: DirectionEvaluationTask

Question 1 (ID: 88ffe47276e19372)

✅ Evaluation Results
id: 88ffe47276e19372
question: You return to your starting position and face north. From a Top-Down map, describe where shelf is relative to truck. Answer format: <cardinal direction>, <distance> Example: north-west, near
action: None
answer: south east, mid distance
task_type: DirectionEvaluationTask
choices: []
kwargs:
(none)
score: 0
evaluation_mode: prompt_cogmap
Evaluation state
Q1: DirectionEvaluationTask

Question 2 (ID: 717cb6885cdcac87)

✅ Evaluation Results
id: 717cb6885cdcac87
question: You return to your starting position and face north. From a Top-Down map, describe where rubberduck is relative to shelf. Answer format: <cardinal direction>, <distance> Example: north-west, near
action: None
answer: north west, far
task_type: DirectionEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: prompt_cogmap
Evaluation state
Q2: DirectionEvaluationTask

Question 3 (ID: c7a5d595404c5cc8)

✅ Evaluation Results
id: c7a5d595404c5cc8
question: You return to your starting position and face north. From a Top-Down map, describe where backpack is relative to pan. Answer format: <cardinal direction>, <distance> Example: north-west, near
action: None
answer: north west, far
task_type: DirectionEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: prompt_cogmap
Evaluation state
Q3: DirectionEvaluationTask

📊 Task: PovEvaluationTask

Question 1 (ID: 14625b47c376defb)

✅ Evaluation Results
id: 14625b47c376defb
question: Now you jump to backpack's direction, facing its direction. Describe where chair is relative to you. Answer format: <ego direction>, <distance> Example: front-left, near
action: None
answer: front-left, mid distance
task_type: PovEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: prompt_cogmap
Evaluation state
Q1: PovEvaluationTask

Question 2 (ID: f49d98e66552e13e)

✅ Evaluation Results
id: f49d98e66552e13e
question: Now you jump to laptop's direction, facing its direction. Describe where shelf is relative to you. Answer format: <ego direction>, <distance> Example: front-left, near
action: None
answer: front, mid distance
task_type: PovEvaluationTask
choices: []
kwargs:
(none)
score: 0.500
evaluation_mode: prompt_cogmap
Evaluation state
Q2: PovEvaluationTask

Question 3 (ID: a211c0863d2dc523)

✅ Evaluation Results
id: a211c0863d2dc523
question: Now you jump to television's direction, facing its direction. Describe where cap is relative to you. Answer format: <ego direction>, <distance> Example: front-left, near
action: None
answer: front, near
task_type: PovEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: prompt_cogmap
Evaluation state
Q3: PovEvaluationTask

📊 Task: BackwardPovTextEvaluationTask

Question 1 (ID: 6f2581c5107b62fa)

✅ Evaluation Results
id: 6f2581c5107b62fa
question: Now you jump to an object's position, facing its direction. You observe that truck is front-left, mid distance, facing backward; shelf is front, mid distance. Which object are you standing at? Answer format: <object_name> Example: lamp
action: None
answer:
answer: laptop
final_pos: [9, 2]
final_ori: [1, 0]
final_observation: [{'name': 'truck', 'direction': 'front-left', 'distance': 'mid distance', 'orientation': 'facing backward'}, {'name': 'shelf', 'direction': 'front', 'distance': 'mid distance', 'orientation': None}]
object_positions:
television: [3.0, 2.0]
chair: [3.0, 10.0]
truck: [11.0, 4.0]
bike: [3.0, 6.0]
backpack: [1.0, 12.0]
pan: [9.0, 4.0]
cap: [5.0, 2.0]
lamp: [1.0, 7.0]
vase: [1.0, 14.0]
shelf: [13.0, 2.0]
rubberduck: [5.0, 14.0]
laptop: [9.0, 2.0]
green door: [7.0, 6.0]
blue door: [6.0, 8.0]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
room_id: 3
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: BackwardPovTextEvaluationTask
choices: []
kwargs:
(none)
name_match: True
view_match: True
score: 1
evaluation_mode: prompt_cogmap
Evaluation state
Q1: BackwardPovTextEvaluationTask

Question 2 (ID: 534ed0980f85c17a)

✅ Evaluation Results
id: 534ed0980f85c17a
question: Now you jump to an object's position, facing its direction. You observe that pan is front, near, facing backward; laptop is front-left, mid distance, facing backward; green door is front-right, slightly far, on front wall. Which object are you standing at? Answer format: <object_name> Example: lamp
action: None
answer:
answer: truck
final_pos: [11, 4]
final_ori: [-1, 0]
final_observation: [{'name': 'pan', 'direction': 'front', 'distance': 'near', 'orientation': 'facing backward'}, {'name': 'laptop', 'direction': 'front-left', 'distance': 'mid distance', 'orientation': 'facing backward'}, {'name': 'green door', 'direction': 'front-right', 'distance': 'slightly far', 'orientation': 'on front wall'}]
object_positions:
television: [3.0, 2.0]
chair: [3.0, 10.0]
truck: [11.0, 4.0]
bike: [3.0, 6.0]
backpack: [1.0, 12.0]
pan: [9.0, 4.0]
cap: [5.0, 2.0]
lamp: [1.0, 7.0]
vase: [1.0, 14.0]
shelf: [13.0, 2.0]
rubberduck: [5.0, 14.0]
laptop: [9.0, 2.0]
green door: [7.0, 6.0]
blue door: [6.0, 8.0]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
room_id: 3
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: BackwardPovTextEvaluationTask
choices: []
kwargs:
(none)
name_match: True
view_match: True
score: 1
evaluation_mode: prompt_cogmap
Evaluation state
Q2: BackwardPovTextEvaluationTask

Question 3 (ID: 947e7c0517a89154)

✅ Evaluation Results
id: 947e7c0517a89154
question: Now you jump to an object's position, facing its direction. You observe that truck is front, near, facing backward; shelf is front-right, slightly far. Which object are you standing at? Answer format: <object_name> Example: lamp
action: None
answer:
answer: pan
final_pos: [9, 4]
final_ori: [1, 0]
final_observation: [{'name': 'truck', 'direction': 'front', 'distance': 'near', 'orientation': 'facing backward'}, {'name': 'shelf', 'direction': 'front-right', 'distance': 'slightly far', 'orientation': None}]
object_positions:
television: [3.0, 2.0]
chair: [3.0, 10.0]
truck: [11.0, 4.0]
bike: [3.0, 6.0]
backpack: [1.0, 12.0]
pan: [9.0, 4.0]
cap: [5.0, 2.0]
lamp: [1.0, 7.0]
vase: [1.0, 14.0]
shelf: [13.0, 2.0]
rubberduck: [5.0, 14.0]
laptop: [9.0, 2.0]
green door: [7.0, 6.0]
blue door: [6.0, 8.0]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
room_id: 3
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: BackwardPovTextEvaluationTask
choices: []
kwargs:
(none)
name_match: True
view_match: True
score: 1
evaluation_mode: prompt_cogmap
Evaluation state
Q3: BackwardPovTextEvaluationTask

📊 Task: Action2ViewEvaluationTask

Question 1 (ID: 4f8a7d910b60df45)

✅ Evaluation Results
id: 4f8a7d910b60df45
question: You return to your starting position and face north. You will execute the following action sequence: 1. Jump to the object at front-right, mid distance. 2. Rotate(-90) 3. Jump to the object at front-right, mid distance. 4. Rotate(-180) After executing the actions, what is the ego relation of bike relative to you? Answer format: <ego direction>, <distance> Example: front, near
action: None
answer: front-right, mid distance
task_type: Action2ViewEvaluationTask
choices: []
kwargs:
(none)
score: 0
evaluation_mode: prompt_cogmap
Evaluation state
Q1: Action2ViewEvaluationTask

Question 2 (ID: 32029f3c5e3c31ee)

✅ Evaluation Results
id: 32029f3c5e3c31ee
question: You return to your starting position and face north. You will execute the following action sequence: 1. Rotate(90) 2. Among objects which are front-left, slightly far to you, you jump to the 2nd from left also nearest one. 3. Among objects which are front-right, slightly far to you, you jump to the 2nd from left also farthest one. 4. Rotate(-180) After executing the actions, what is the ego relation of truck relative to you? Answer format: <ego direction>, <distance> Example: front, near
action: None
answer: front-right, mid distance
task_type: Action2ViewEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: prompt_cogmap
Evaluation state
Q2: Action2ViewEvaluationTask

Question 3 (ID: 79787b772d0b8410)

✅ Evaluation Results
id: 79787b772d0b8410
question: You return to your starting position and face north. You will execute the following action sequence: 1. Jump to the object at front-right, mid distance. 2. Rotate(90) 3. Jump to the object at front-left, mid distance. 4. Rotate(90) After executing the actions, what is the ego relation of cap relative to you? Answer format: <ego direction>, <distance> Example: front, near
action: None
answer: front-slight-right, slightly far
task_type: Action2ViewEvaluationTask
choices: []
kwargs:
(none)
score: 0
evaluation_mode: prompt_cogmap
Evaluation state
Q3: Action2ViewEvaluationTask

📊 Task: View2ActionTextEvaluationTask

Question 1 (ID: a13a817826736b87)

✅ Evaluation Results
id: a13a817826736b87
question: You return to your starting position and face north. Then you have executed an action sequence and changed to a new location and facing direction. You observe the following: pan is at front-right, slightly far, facing backward; truck is at front-right, mid distance, facing forward; laptop is at front, mid distance, facing backward What action sequence led to this final view? The action sequence must be valid and only contain move actions. Answer format: <sequence of move actions> Example: JumpTo(lamp), Rotate(90)
action: None
answer:
final_pos: [13, 2]
final_ori: [-1, 0]
room_id: 3
init_pos: [1, 3]
init_ori: [0, 1]
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
minimal_plan: [['rotate', 90], ['jumpto', 'green door'], ['jumpto', 'shelf'], ['rotate', 180]]
final_observation: [{'name': 'pan', 'direction': 'front-right', 'distance': 'slightly far', 'orientation': 'facing backward'}, {'name': 'truck', 'direction': 'front-right', 'distance': 'mid distance', 'orientation': 'facing forward'}, {'name': 'laptop', 'direction': 'front', 'distance': 'mid distance', 'orientation': 'facing backward'}]
task_type: View2ActionTextEvaluationTask
choices: []
kwargs:
(none)
error: invalid_format
score: 0
evaluation_mode: prompt_cogmap
Evaluation state
Q1: View2ActionTextEvaluationTask

Question 2 (ID: ef2bc20cb1343cb5)

✅ Evaluation Results
id: ef2bc20cb1343cb5
question: You return to your starting position and face north. Then you have executed an action sequence and changed to a new location and facing direction. You observe the following: blue door is at front-left, mid distance, on front wall What action sequence led to this final view? The action sequence must be valid and only contain move actions. Answer format: <sequence of move actions> Example: JumpTo(lamp), Rotate(90)
action: None
answer:
final_pos: [7, 6]
final_ori: [0, 1]
room_id: [1, 3]
init_pos: [1, 3]
init_ori: [0, 1]
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
minimal_plan: [['rotate', 90], ['jumpto', 'initial_pos'], ['rotate', -90]]
final_observation: [{'name': 'blue door', 'direction': 'front-left', 'distance': 'mid distance', 'orientation': 'on front wall'}]
task_type: View2ActionTextEvaluationTask
choices: []
kwargs:
(none)
pos_match: False
ori_match: True
final_pos: [3.0, 6.0]
final_ori: [0, 1]
visible_match: False
score: 0
evaluation_mode: prompt_cogmap
Evaluation state
Q2: View2ActionTextEvaluationTask

Question 3 (ID: 3d92d2ea666b2ed5)

✅ Evaluation Results
id: 3d92d2ea666b2ed5
question: You return to your starting position and face north. Then you have executed an action sequence and changed to a new location and facing direction. You observe the following: laptop is at front-left, slightly far, facing left; cap is at front-right, slightly far, facing left; pan is at front-left, mid distance, facing left What action sequence led to this final view? The action sequence must be valid and only contain move actions. Answer format: <sequence of move actions> Example: JumpTo(lamp), Rotate(90)
action: None
answer:
final_pos: [7, 6]
final_ori: [0, -1]
room_id: [1, 3]
init_pos: [1, 3]
init_ori: [0, 1]
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
minimal_plan: [['rotate', 90], ['jumpto', 'initial_pos'], ['rotate', 90]]
final_observation: [{'name': 'laptop', 'direction': 'front-left', 'distance': 'slightly far', 'orientation': 'facing left'}, {'name': 'cap', 'direction': 'front-right', 'distance': 'slightly far', 'orientation': 'facing left'}, {'name': 'pan', 'direction': 'front-left', 'distance': 'mid distance', 'orientation': 'facing left'}]
task_type: View2ActionTextEvaluationTask
choices: []
kwargs:
(none)
pos_match: True
ori_match: True
final_pos: [7.0, 6.0]
final_ori: [0, -1]
visible_match: True
score: 1
evaluation_mode: prompt_cogmap
Evaluation state
Q3: View2ActionTextEvaluationTask

📊 Task: AlloMappingEvaluationTask

Question 1 (ID: 47868a3a901e547c)

✅ Evaluation Results
id: 47868a3a901e547c
question: Treat your starting position as the origin (0, 0) while facing north. Report allocentric coordinates using (x right/east, y up/north). Objects: shelf, truck, lamp. Answer format: (x0, y0); (x1, y1); ... in the same order. Example: (1, 0); (-2, 3); (0, -1)
action: None
answer: [[12, -1], [10, 1], [0, 4]]
task_type: AlloMappingEvaluationTask
choices: []
kwargs:
(none)
similarity: 0.779
threshold: 0.900
score: 0.779
evaluation_mode: prompt_cogmap
Evaluation state
Q1: AlloMappingEvaluationTask

Question 2 (ID: adbcc906c717830a)

✅ Evaluation Results
id: adbcc906c717830a
question: Treat your starting position as the origin (0, 0) while facing north. Report allocentric coordinates using (x right/east, y up/north). Objects: rubberduck, shelf, pan. Answer format: (x0, y0); (x1, y1); ... in the same order. Example: (1, 0); (-2, 3); (0, -1)
action: None
answer: [[4, 11], [12, -1], [8, 1]]
task_type: AlloMappingEvaluationTask
choices: []
kwargs:
(none)
similarity: 0.820
threshold: 0.900
score: 0.820
evaluation_mode: prompt_cogmap
Evaluation state
Q2: AlloMappingEvaluationTask

Question 3 (ID: 9cdfd701876daa0f)

✅ Evaluation Results
id: 9cdfd701876daa0f
question: Treat your starting position as the origin (0, 0) while facing north. Report allocentric coordinates using (x right/east, y up/north). Objects: cap, pan, backpack, shelf. Answer format: (x0, y0); (x1, y1); ... in the same order. Example: (1, 0); (-2, 3); (0, -1)
action: None
answer: [[4, -1], [8, 1], [0, 9], [12, -1]]
task_type: AlloMappingEvaluationTask
choices: []
kwargs:
(none)
similarity: 0.760
threshold: 0.900
score: 0.760
evaluation_mode: prompt_cogmap
Evaluation state
Q3: AlloMappingEvaluationTask

📊 Task: RotEvaluationTask

Question 1 (ID: c7b0da8750e06c67)

✅ Evaluation Results
id: c7b0da8750e06c67
question: You return to your starting position and face north. You will perform a full 360-degree rotation by continuously turning counterclockwise in place. Assume all walls are removed (you can see through walls), so every object is visible. Focus on this set of objects: bike, pan, television. List them in the exact order they appear directly ahead while you rotate. If two objects share a bearing, place the nearer one first. Answer format: <object_name1>, <object_name2>, ... Example: mug, sofa, plant
action: None
answer: ['television', 'pan', 'bike']
task_type: RotEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: prompt_cogmap
Evaluation state
Q1: RotEvaluationTask

Question 2 (ID: fc92fab04fedd039)

✅ Evaluation Results
id: fc92fab04fedd039
question: You return to your starting position and face north. You will perform a full 360-degree rotation by continuously turning counterclockwise in place. Assume all walls are removed (you can see through walls), so every object is visible. Focus on this set of objects: bike, television, truck. List them in the exact order they appear directly ahead while you rotate. If two objects share a bearing, place the nearer one first. Answer format: <object_name1>, <object_name2>, ... Example: mug, sofa, plant
action: None
answer: ['television', 'truck', 'bike']
task_type: RotEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: prompt_cogmap
Evaluation state
Q2: RotEvaluationTask

Question 3 (ID: 83da2dae108d2082)

✅ Evaluation Results
id: 83da2dae108d2082
question: You return to your starting position and face north. You will perform a full 360-degree rotation by continuously turning counterclockwise in place. Assume all walls are removed (you can see through walls), so every object is visible. Focus on this set of objects: chair, television, truck. List them in the exact order they appear directly ahead while you rotate. If two objects share a bearing, place the nearer one first. Answer format: <object_name1>, <object_name2>, ... Example: mug, sofa, plant
action: None
answer: ['television', 'truck', 'chair']
task_type: RotEvaluationTask
choices: []
kwargs:
(none)
score: 0
evaluation_mode: prompt_cogmap
Evaluation state
Q3: RotEvaluationTask

📊 Task: Location2ViewEvaluationTask

Question 1 (ID: 13d7589043fe1ee0)

✅ Evaluation Results
id: 13d7589043fe1ee0
question: Treat the green door as the new 'origin' (0, 0). You move to (2, -5) and face north. What is the egocentric relation of pan? Answer format: <direction>, <distance> Example: front, near
action: None
answer: front, mid distance
task_type: Location2ViewEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: prompt_cogmap
Evaluation state
Q1: Location2ViewEvaluationTask

Question 2 (ID: e2c7ab6a2049c06a)

✅ Evaluation Results
id: e2c7ab6a2049c06a
question: Still treat your initial position as origin (0, 0) You move to (4, 4) and face west. What is the egocentric relation of bike? Answer format: <direction>, <distance> Example: front, near
action: None
answer: front-left, mid distance
task_type: Location2ViewEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: prompt_cogmap
Evaluation state
Q2: Location2ViewEvaluationTask

Question 3 (ID: 5a56a83a64962b71)

✅ Evaluation Results
id: 5a56a83a64962b71
question: Still treat your initial position as origin (0, 0) You move to (0, 3) and face east. What is the egocentric relation of blue door? Answer format: <direction>, <distance> Example: front, near
action: None
answer: front-slight-left, slightly far
task_type: Location2ViewEvaluationTask
choices: []
kwargs:
(none)
score: 0
evaluation_mode: prompt_cogmap
Evaluation state
Q3: Location2ViewEvaluationTask

📊 Task: View2LocationTextEvaluationTask

Question 1 (ID: 19a422deab9e39fc)

✅ Evaluation Results
id: 19a422deab9e39fc
question: You move to a new location and face north. You observe: pan is at front, mid distance, facing right; truck is at front-right, mid distance, facing left; green door is at front-slight-left, slightly far, on left wall Treat the green door as the new 'origin' (0, 0). What is your new 2D coordinate (x, y)? Answer format: (x, y) Example: (2, -1)
action: None
answer:
coord: [2, -5]
final_pos: [9, 1]
final_ori: [0, 1]
room_id: 3
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
final_observation: [{'name': 'pan', 'direction': 'front', 'distance': 'mid distance', 'orientation': 'facing right'}, {'name': 'truck', 'direction': 'front-right', 'distance': 'mid distance', 'orientation': 'facing left'}, {'name': 'green door', 'direction': 'front-slight-left', 'distance': 'slightly far', 'orientation': 'on left wall'}]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: View2LocationTextEvaluationTask
choices: []
kwargs:
(none)
similarity: 1
raw_score: 1
score: 1
evaluation_mode: prompt_cogmap
Evaluation state
Q1: View2LocationTextEvaluationTask

Question 2 (ID: 197270037eaa3b24)

✅ Evaluation Results
id: 197270037eaa3b24
question: You move to a new location and face west. You observe: bike is at front-left, mid distance, facing forward; lamp is at front, mid distance Still treat your initial position as origin (0, 0) What is your new 2D coordinate (x, y)? Answer format: (x, y) Example: (2, -1)
action: None
answer:
coord: [4, 4]
final_pos: [5, 7]
final_ori: [-1, 0]
room_id: 1
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
final_observation: [{'name': 'bike', 'direction': 'front-left', 'distance': 'mid distance', 'orientation': 'facing forward'}, {'name': 'lamp', 'direction': 'front', 'distance': 'mid distance', 'orientation': None}]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: View2LocationTextEvaluationTask
choices: []
kwargs:
(none)
similarity: 0.779
raw_score: 0.779
score: 0.779
evaluation_mode: prompt_cogmap
Evaluation state
Q2: View2LocationTextEvaluationTask

Question 3 (ID: 7d2a51a882153983)

✅ Evaluation Results
id: 7d2a51a882153983
question: You move to a new location and face east. You observe: blue door is at front-slight-left, slightly far, on left wall; cap is at front-right, slightly far, facing forward; bike is at front, near, facing backward Still treat your initial position as origin (0, 0) What is your new 2D coordinate (x, y)? Answer format: (x, y) Example: (2, -1)
action: None
answer:
coord: [0, 3]
final_pos: [1, 6]
final_ori: [1, 0]
room_id: 1
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
final_observation: [{'name': 'blue door', 'direction': 'front-slight-left', 'distance': 'slightly far', 'orientation': 'on left wall'}, {'name': 'cap', 'direction': 'front-right', 'distance': 'slightly far', 'orientation': 'facing forward'}, {'name': 'bike', 'direction': 'front', 'distance': 'near', 'orientation': 'facing backward'}]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: View2LocationTextEvaluationTask
choices: []
kwargs:
(none)
similarity: 0.624
raw_score: 0.624
score: 0.624
evaluation_mode: prompt_cogmap
Evaluation state
Q3: View2LocationTextEvaluationTask

📊 Evaluation (use_gt_cogmap)

📊 Task: RotEvaluationTask

Question 1 (ID: fc92fab04fedd039)

✅ Evaluation Results
id: fc92fab04fedd039
question: You return to your starting position and face north. You will perform a full 360-degree rotation by continuously turning counterclockwise in place. Assume all walls are removed (you can see through walls), so every object is visible. Focus on this set of objects: bike, television, truck. List them in the exact order they appear directly ahead while you rotate. If two objects share a bearing, place the nearer one first. Answer format: <object_name1>, <object_name2>, ... Example: mug, sofa, plant
action: None
answer: ['television', 'truck', 'bike']
task_type: RotEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q1: RotEvaluationTask

Question 2 (ID: 83da2dae108d2082)

✅ Evaluation Results
id: 83da2dae108d2082
question: You return to your starting position and face north. You will perform a full 360-degree rotation by continuously turning counterclockwise in place. Assume all walls are removed (you can see through walls), so every object is visible. Focus on this set of objects: chair, television, truck. List them in the exact order they appear directly ahead while you rotate. If two objects share a bearing, place the nearer one first. Answer format: <object_name1>, <object_name2>, ... Example: mug, sofa, plant
action: None
answer: ['television', 'truck', 'chair']
task_type: RotEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q2: RotEvaluationTask

Question 3 (ID: c7b0da8750e06c67)

✅ Evaluation Results
id: c7b0da8750e06c67
question: You return to your starting position and face north. You will perform a full 360-degree rotation by continuously turning counterclockwise in place. Assume all walls are removed (you can see through walls), so every object is visible. Focus on this set of objects: bike, pan, television. List them in the exact order they appear directly ahead while you rotate. If two objects share a bearing, place the nearer one first. Answer format: <object_name1>, <object_name2>, ... Example: mug, sofa, plant
action: None
answer: ['television', 'pan', 'bike']
task_type: RotEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q3: RotEvaluationTask

📊 Task: Location2ViewEvaluationTask

Question 1 (ID: 13d7589043fe1ee0)

✅ Evaluation Results
id: 13d7589043fe1ee0
question: Treat the green door as the new 'origin' (0, 0). You move to (2, -5) and face north. What is the egocentric relation of pan? Answer format: <direction>, <distance> Example: front, near
action: None
answer: front, mid distance
task_type: Location2ViewEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q1: Location2ViewEvaluationTask

Question 2 (ID: e2c7ab6a2049c06a)

✅ Evaluation Results
id: e2c7ab6a2049c06a
question: Still treat your initial position as origin (0, 0) You move to (4, 4) and face west. What is the egocentric relation of bike? Answer format: <direction>, <distance> Example: front, near
action: None
answer: front-left, mid distance
task_type: Location2ViewEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q2: Location2ViewEvaluationTask

Question 3 (ID: 5a56a83a64962b71)

✅ Evaluation Results
id: 5a56a83a64962b71
question: Still treat your initial position as origin (0, 0) You move to (0, 3) and face east. What is the egocentric relation of blue door? Answer format: <direction>, <distance> Example: front, near
action: None
answer: front-slight-left, slightly far
task_type: Location2ViewEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q3: Location2ViewEvaluationTask

📊 Task: View2LocationTextEvaluationTask

Question 1 (ID: 19a422deab9e39fc)

✅ Evaluation Results
id: 19a422deab9e39fc
question: You move to a new location and face north. You observe: pan is at front, mid distance, facing right; truck is at front-right, mid distance, facing left; green door is at front-slight-left, slightly far, on left wall Treat the green door as the new 'origin' (0, 0). What is your new 2D coordinate (x, y)? Answer format: (x, y) Example: (2, -1)
action: None
answer:
coord: [2, -5]
final_pos: [9, 1]
final_ori: [0, 1]
room_id: 3
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
final_observation: [{'name': 'pan', 'direction': 'front', 'distance': 'mid distance', 'orientation': 'facing right'}, {'name': 'truck', 'direction': 'front-right', 'distance': 'mid distance', 'orientation': 'facing left'}, {'name': 'green door', 'direction': 'front-slight-left', 'distance': 'slightly far', 'orientation': 'on left wall'}]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: View2LocationTextEvaluationTask
choices: []
kwargs:
(none)
similarity: 1
raw_score: 1
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q1: View2LocationTextEvaluationTask

Question 2 (ID: 197270037eaa3b24)

✅ Evaluation Results
id: 197270037eaa3b24
question: You move to a new location and face west. You observe: bike is at front-left, mid distance, facing forward; lamp is at front, mid distance Still treat your initial position as origin (0, 0) What is your new 2D coordinate (x, y)? Answer format: (x, y) Example: (2, -1)
action: None
answer:
coord: [4, 4]
final_pos: [5, 7]
final_ori: [-1, 0]
room_id: 1
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
final_observation: [{'name': 'bike', 'direction': 'front-left', 'distance': 'mid distance', 'orientation': 'facing forward'}, {'name': 'lamp', 'direction': 'front', 'distance': 'mid distance', 'orientation': None}]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: View2LocationTextEvaluationTask
choices: []
kwargs:
(none)
similarity: 1
raw_score: 1
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q2: View2LocationTextEvaluationTask

Question 3 (ID: 7d2a51a882153983)

✅ Evaluation Results
id: 7d2a51a882153983
question: You move to a new location and face east. You observe: blue door is at front-slight-left, slightly far, on left wall; cap is at front-right, slightly far, facing forward; bike is at front, near, facing backward Still treat your initial position as origin (0, 0) What is your new 2D coordinate (x, y)? Answer format: (x, y) Example: (2, -1)
action: None
answer:
coord: [0, 3]
final_pos: [1, 6]
final_ori: [1, 0]
room_id: 1
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
final_observation: [{'name': 'blue door', 'direction': 'front-slight-left', 'distance': 'slightly far', 'orientation': 'on left wall'}, {'name': 'cap', 'direction': 'front-right', 'distance': 'slightly far', 'orientation': 'facing forward'}, {'name': 'bike', 'direction': 'front', 'distance': 'near', 'orientation': 'facing backward'}]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: View2LocationTextEvaluationTask
choices: []
kwargs:
(none)
similarity: 1
raw_score: 1
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q3: View2LocationTextEvaluationTask

📊 Task: DirectionEvaluationTask

Question 1 (ID: 88ffe47276e19372)

✅ Evaluation Results
id: 88ffe47276e19372
question: You return to your starting position and face north. From a Top-Down map, describe where shelf is relative to truck. Answer format: <cardinal direction>, <distance> Example: north-west, near
action: None
answer: south east, mid distance
task_type: DirectionEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q1: DirectionEvaluationTask

Question 2 (ID: 717cb6885cdcac87)

✅ Evaluation Results
id: 717cb6885cdcac87
question: You return to your starting position and face north. From a Top-Down map, describe where rubberduck is relative to shelf. Answer format: <cardinal direction>, <distance> Example: north-west, near
action: None
answer: north west, far
task_type: DirectionEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q2: DirectionEvaluationTask

Question 3 (ID: c7a5d595404c5cc8)

✅ Evaluation Results
id: c7a5d595404c5cc8
question: You return to your starting position and face north. From a Top-Down map, describe where backpack is relative to pan. Answer format: <cardinal direction>, <distance> Example: north-west, near
action: None
answer: north west, far
task_type: DirectionEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q3: DirectionEvaluationTask

📊 Task: PovEvaluationTask

Question 1 (ID: 14625b47c376defb)

✅ Evaluation Results
id: 14625b47c376defb
question: Now you jump to backpack's direction, facing its direction. Describe where chair is relative to you. Answer format: <ego direction>, <distance> Example: front-left, near
action: None
answer: front-left, mid distance
task_type: PovEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q1: PovEvaluationTask

Question 2 (ID: f49d98e66552e13e)

✅ Evaluation Results
id: f49d98e66552e13e
question: Now you jump to laptop's direction, facing its direction. Describe where shelf is relative to you. Answer format: <ego direction>, <distance> Example: front-left, near
action: None
answer: front, mid distance
task_type: PovEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q2: PovEvaluationTask

Question 3 (ID: a211c0863d2dc523)

✅ Evaluation Results
id: a211c0863d2dc523
question: Now you jump to television's direction, facing its direction. Describe where cap is relative to you. Answer format: <ego direction>, <distance> Example: front-left, near
action: None
answer: front, near
task_type: PovEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q3: PovEvaluationTask

📊 Task: BackwardPovTextEvaluationTask

Question 1 (ID: 6f2581c5107b62fa)

✅ Evaluation Results
id: 6f2581c5107b62fa
question: Now you jump to an object's position, facing its direction. You observe that truck is front-left, mid distance, facing backward; shelf is front, mid distance. Which object are you standing at? Answer format: <object_name> Example: lamp
action: None
answer:
answer: laptop
final_pos: [9, 2]
final_ori: [1, 0]
final_observation: [{'name': 'truck', 'direction': 'front-left', 'distance': 'mid distance', 'orientation': 'facing backward'}, {'name': 'shelf', 'direction': 'front', 'distance': 'mid distance', 'orientation': None}]
object_positions:
television: [3.0, 2.0]
chair: [3.0, 10.0]
truck: [11.0, 4.0]
bike: [3.0, 6.0]
backpack: [1.0, 12.0]
pan: [9.0, 4.0]
cap: [5.0, 2.0]
lamp: [1.0, 7.0]
vase: [1.0, 14.0]
shelf: [13.0, 2.0]
rubberduck: [5.0, 14.0]
laptop: [9.0, 2.0]
green door: [7.0, 6.0]
blue door: [6.0, 8.0]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
room_id: 3
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: BackwardPovTextEvaluationTask
choices: []
kwargs:
(none)
name_match: True
view_match: True
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q1: BackwardPovTextEvaluationTask

Question 2 (ID: 534ed0980f85c17a)

✅ Evaluation Results
id: 534ed0980f85c17a
question: Now you jump to an object's position, facing its direction. You observe that pan is front, near, facing backward; laptop is front-left, mid distance, facing backward; green door is front-right, slightly far, on front wall. Which object are you standing at? Answer format: <object_name> Example: lamp
action: None
answer:
answer: truck
final_pos: [11, 4]
final_ori: [-1, 0]
final_observation: [{'name': 'pan', 'direction': 'front', 'distance': 'near', 'orientation': 'facing backward'}, {'name': 'laptop', 'direction': 'front-left', 'distance': 'mid distance', 'orientation': 'facing backward'}, {'name': 'green door', 'direction': 'front-right', 'distance': 'slightly far', 'orientation': 'on front wall'}]
object_positions:
television: [3.0, 2.0]
chair: [3.0, 10.0]
truck: [11.0, 4.0]
bike: [3.0, 6.0]
backpack: [1.0, 12.0]
pan: [9.0, 4.0]
cap: [5.0, 2.0]
lamp: [1.0, 7.0]
vase: [1.0, 14.0]
shelf: [13.0, 2.0]
rubberduck: [5.0, 14.0]
laptop: [9.0, 2.0]
green door: [7.0, 6.0]
blue door: [6.0, 8.0]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
room_id: 3
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: BackwardPovTextEvaluationTask
choices: []
kwargs:
(none)
name_match: True
view_match: True
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q2: BackwardPovTextEvaluationTask

Question 3 (ID: 947e7c0517a89154)

✅ Evaluation Results
id: 947e7c0517a89154
question: Now you jump to an object's position, facing its direction. You observe that truck is front, near, facing backward; shelf is front-right, slightly far. Which object are you standing at? Answer format: <object_name> Example: lamp
action: None
answer:
answer: pan
final_pos: [9, 4]
final_ori: [1, 0]
final_observation: [{'name': 'truck', 'direction': 'front', 'distance': 'near', 'orientation': 'facing backward'}, {'name': 'shelf', 'direction': 'front-right', 'distance': 'slightly far', 'orientation': None}]
object_positions:
television: [3.0, 2.0]
chair: [3.0, 10.0]
truck: [11.0, 4.0]
bike: [3.0, 6.0]
backpack: [1.0, 12.0]
pan: [9.0, 4.0]
cap: [5.0, 2.0]
lamp: [1.0, 7.0]
vase: [1.0, 14.0]
shelf: [13.0, 2.0]
rubberduck: [5.0, 14.0]
laptop: [9.0, 2.0]
green door: [7.0, 6.0]
blue door: [6.0, 8.0]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
room_id: 3
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: BackwardPovTextEvaluationTask
choices: []
kwargs:
(none)
name_match: True
view_match: True
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q3: BackwardPovTextEvaluationTask

📊 Task: Action2ViewEvaluationTask

Question 1 (ID: 4f8a7d910b60df45)

✅ Evaluation Results
id: 4f8a7d910b60df45
question: You return to your starting position and face north. You will execute the following action sequence: 1. Jump to the object at front-right, mid distance. 2. Rotate(-90) 3. Jump to the object at front-right, mid distance. 4. Rotate(-180) After executing the actions, what is the ego relation of bike relative to you? Answer format: <ego direction>, <distance> Example: front, near
action: None
answer: front-right, mid distance
task_type: Action2ViewEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q1: Action2ViewEvaluationTask

Question 2 (ID: 32029f3c5e3c31ee)

✅ Evaluation Results
id: 32029f3c5e3c31ee
question: You return to your starting position and face north. You will execute the following action sequence: 1. Rotate(90) 2. Among objects which are front-left, slightly far to you, you jump to the 2nd from left also nearest one. 3. Among objects which are front-right, slightly far to you, you jump to the 2nd from left also farthest one. 4. Rotate(-180) After executing the actions, what is the ego relation of truck relative to you? Answer format: <ego direction>, <distance> Example: front, near
action: None
answer: front-right, mid distance
task_type: Action2ViewEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q2: Action2ViewEvaluationTask

Question 3 (ID: 79787b772d0b8410)

✅ Evaluation Results
id: 79787b772d0b8410
question: You return to your starting position and face north. You will execute the following action sequence: 1. Jump to the object at front-right, mid distance. 2. Rotate(90) 3. Jump to the object at front-left, mid distance. 4. Rotate(90) After executing the actions, what is the ego relation of cap relative to you? Answer format: <ego direction>, <distance> Example: front, near
action: None
answer: front-slight-right, slightly far
task_type: Action2ViewEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q3: Action2ViewEvaluationTask

📊 Task: View2ActionTextEvaluationTask

Question 1 (ID: a13a817826736b87)

✅ Evaluation Results
id: a13a817826736b87
question: You return to your starting position and face north. Then you have executed an action sequence and changed to a new location and facing direction. You observe the following: pan is at front-right, slightly far, facing backward; truck is at front-right, mid distance, facing forward; laptop is at front, mid distance, facing backward What action sequence led to this final view? The action sequence must be valid and only contain move actions. Answer format: <sequence of move actions> Example: JumpTo(lamp), Rotate(90)
action: None
answer:
final_pos: [13, 2]
final_ori: [-1, 0]
room_id: 3
init_pos: [1, 3]
init_ori: [0, 1]
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
minimal_plan: [['rotate', 90], ['jumpto', 'green door'], ['jumpto', 'shelf'], ['rotate', 180]]
final_observation: [{'name': 'pan', 'direction': 'front-right', 'distance': 'slightly far', 'orientation': 'facing backward'}, {'name': 'truck', 'direction': 'front-right', 'distance': 'mid distance', 'orientation': 'facing forward'}, {'name': 'laptop', 'direction': 'front', 'distance': 'mid distance', 'orientation': 'facing backward'}]
task_type: View2ActionTextEvaluationTask
choices: []
kwargs:
(none)
pos_match: True
ori_match: True
final_pos: [13.0, 2.0]
final_ori: [-1, 0]
visible_match: True
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q1: View2ActionTextEvaluationTask

Question 2 (ID: ef2bc20cb1343cb5)

✅ Evaluation Results
id: ef2bc20cb1343cb5
question: You return to your starting position and face north. Then you have executed an action sequence and changed to a new location and facing direction. You observe the following: blue door is at front-left, mid distance, on front wall What action sequence led to this final view? The action sequence must be valid and only contain move actions. Answer format: <sequence of move actions> Example: JumpTo(lamp), Rotate(90)
action: None
answer:
final_pos: [7, 6]
final_ori: [0, 1]
room_id: [1, 3]
init_pos: [1, 3]
init_ori: [0, 1]
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
minimal_plan: [['rotate', 90], ['jumpto', 'initial_pos'], ['rotate', -90]]
final_observation: [{'name': 'blue door', 'direction': 'front-left', 'distance': 'mid distance', 'orientation': 'on front wall'}]
task_type: View2ActionTextEvaluationTask
choices: []
kwargs:
(none)
error: invalid_format
score: 0
evaluation_mode: use_gt_cogmap
Evaluation state
Q2: View2ActionTextEvaluationTask

Question 3 (ID: 3d92d2ea666b2ed5)

✅ Evaluation Results
id: 3d92d2ea666b2ed5
question: You return to your starting position and face north. Then you have executed an action sequence and changed to a new location and facing direction. You observe the following: laptop is at front-left, slightly far, facing left; cap is at front-right, slightly far, facing left; pan is at front-left, mid distance, facing left What action sequence led to this final view? The action sequence must be valid and only contain move actions. Answer format: <sequence of move actions> Example: JumpTo(lamp), Rotate(90)
action: None
answer:
final_pos: [7, 6]
final_ori: [0, -1]
room_id: [1, 3]
init_pos: [1, 3]
init_ori: [0, 1]
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
minimal_plan: [['rotate', 90], ['jumpto', 'initial_pos'], ['rotate', 90]]
final_observation: [{'name': 'laptop', 'direction': 'front-left', 'distance': 'slightly far', 'orientation': 'facing left'}, {'name': 'cap', 'direction': 'front-right', 'distance': 'slightly far', 'orientation': 'facing left'}, {'name': 'pan', 'direction': 'front-left', 'distance': 'mid distance', 'orientation': 'facing left'}]
task_type: View2ActionTextEvaluationTask
choices: []
kwargs:
(none)
error: invalid_format
score: 0
evaluation_mode: use_gt_cogmap
Evaluation state
Q3: View2ActionTextEvaluationTask

📊 Task: AlloMappingEvaluationTask

Question 1 (ID: 47868a3a901e547c)

✅ Evaluation Results
id: 47868a3a901e547c
question: Treat your starting position as the origin (0, 0) while facing north. Report allocentric coordinates using (x right/east, y up/north). Objects: shelf, truck, lamp. Answer format: (x0, y0); (x1, y1); ... in the same order. Example: (1, 0); (-2, 3); (0, -1)
action: None
answer: [[12, -1], [10, 1], [0, 4]]
task_type: AlloMappingEvaluationTask
choices: []
kwargs:
(none)
similarity: 1
threshold: 0.900
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q1: AlloMappingEvaluationTask

Question 2 (ID: adbcc906c717830a)

✅ Evaluation Results
id: adbcc906c717830a
question: Treat your starting position as the origin (0, 0) while facing north. Report allocentric coordinates using (x right/east, y up/north). Objects: rubberduck, shelf, pan. Answer format: (x0, y0); (x1, y1); ... in the same order. Example: (1, 0); (-2, 3); (0, -1)
action: None
answer: [[4, 11], [12, -1], [8, 1]]
task_type: AlloMappingEvaluationTask
choices: []
kwargs:
(none)
similarity: 1
threshold: 0.900
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q2: AlloMappingEvaluationTask

Question 3 (ID: 9cdfd701876daa0f)

✅ Evaluation Results
id: 9cdfd701876daa0f
question: Treat your starting position as the origin (0, 0) while facing north. Report allocentric coordinates using (x right/east, y up/north). Objects: cap, pan, backpack, shelf. Answer format: (x0, y0); (x1, y1); ... in the same order. Example: (1, 0); (-2, 3); (0, -1)
action: None
answer: [[4, -1], [8, 1], [0, 9], [12, -1]]
task_type: AlloMappingEvaluationTask
choices: []
kwargs:
(none)
similarity: 1
threshold: 0.900
score: 1
evaluation_mode: use_gt_cogmap
Evaluation state
Q3: AlloMappingEvaluationTask

📊 Evaluation (use_model_cogmap)

📊 Task: DirectionEvaluationTask

Question 1 (ID: 88ffe47276e19372)

✅ Evaluation Results
id: 88ffe47276e19372
question: You return to your starting position and face north. From a Top-Down map, describe where shelf is relative to truck. Answer format: <cardinal direction>, <distance> Example: north-west, near
action: None
answer: south east, mid distance
task_type: DirectionEvaluationTask
choices: []
kwargs:
(none)
score: 0
evaluation_mode: use_model_cogmap
Evaluation state
Q1: DirectionEvaluationTask

Question 2 (ID: 717cb6885cdcac87)

✅ Evaluation Results
id: 717cb6885cdcac87
question: You return to your starting position and face north. From a Top-Down map, describe where rubberduck is relative to shelf. Answer format: <cardinal direction>, <distance> Example: north-west, near
action: None
answer: north west, far
task_type: DirectionEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_model_cogmap
Evaluation state
Q2: DirectionEvaluationTask

Question 3 (ID: c7a5d595404c5cc8)

✅ Evaluation Results
id: c7a5d595404c5cc8
question: You return to your starting position and face north. From a Top-Down map, describe where backpack is relative to pan. Answer format: <cardinal direction>, <distance> Example: north-west, near
action: None
answer: north west, far
task_type: DirectionEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_model_cogmap
Evaluation state
Q3: DirectionEvaluationTask

📊 Task: PovEvaluationTask

Question 1 (ID: 14625b47c376defb)

✅ Evaluation Results
id: 14625b47c376defb
question: Now you jump to backpack's direction, facing its direction. Describe where chair is relative to you. Answer format: <ego direction>, <distance> Example: front-left, near
action: None
answer: front-left, mid distance
task_type: PovEvaluationTask
choices: []
kwargs:
(none)
score: 0
evaluation_mode: use_model_cogmap
Evaluation state
Q1: PovEvaluationTask

Question 2 (ID: f49d98e66552e13e)

✅ Evaluation Results
id: f49d98e66552e13e
question: Now you jump to laptop's direction, facing its direction. Describe where shelf is relative to you. Answer format: <ego direction>, <distance> Example: front-left, near
action: None
answer: front, mid distance
task_type: PovEvaluationTask
choices: []
kwargs:
(none)
score: 0.500
evaluation_mode: use_model_cogmap
Evaluation state
Q2: PovEvaluationTask

Question 3 (ID: a211c0863d2dc523)

✅ Evaluation Results
id: a211c0863d2dc523
question: Now you jump to television's direction, facing its direction. Describe where cap is relative to you. Answer format: <ego direction>, <distance> Example: front-left, near
action: None
answer: front, near
task_type: PovEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_model_cogmap
Evaluation state
Q3: PovEvaluationTask

📊 Task: BackwardPovTextEvaluationTask

Question 1 (ID: 6f2581c5107b62fa)

✅ Evaluation Results
id: 6f2581c5107b62fa
question: Now you jump to an object's position, facing its direction. You observe that truck is front-left, mid distance, facing backward; shelf is front, mid distance. Which object are you standing at? Answer format: <object_name> Example: lamp
action: None
answer:
answer: laptop
final_pos: [9, 2]
final_ori: [1, 0]
final_observation: [{'name': 'truck', 'direction': 'front-left', 'distance': 'mid distance', 'orientation': 'facing backward'}, {'name': 'shelf', 'direction': 'front', 'distance': 'mid distance', 'orientation': None}]
object_positions:
television: [3.0, 2.0]
chair: [3.0, 10.0]
truck: [11.0, 4.0]
bike: [3.0, 6.0]
backpack: [1.0, 12.0]
pan: [9.0, 4.0]
cap: [5.0, 2.0]
lamp: [1.0, 7.0]
vase: [1.0, 14.0]
shelf: [13.0, 2.0]
rubberduck: [5.0, 14.0]
laptop: [9.0, 2.0]
green door: [7.0, 6.0]
blue door: [6.0, 8.0]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
room_id: 3
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: BackwardPovTextEvaluationTask
choices: []
kwargs:
(none)
name_match: True
view_match: True
score: 1
evaluation_mode: use_model_cogmap
Evaluation state
Q1: BackwardPovTextEvaluationTask

Question 2 (ID: 534ed0980f85c17a)

✅ Evaluation Results
id: 534ed0980f85c17a
question: Now you jump to an object's position, facing its direction. You observe that pan is front, near, facing backward; laptop is front-left, mid distance, facing backward; green door is front-right, slightly far, on front wall. Which object are you standing at? Answer format: <object_name> Example: lamp
action: None
answer:
answer: truck
final_pos: [11, 4]
final_ori: [-1, 0]
final_observation: [{'name': 'pan', 'direction': 'front', 'distance': 'near', 'orientation': 'facing backward'}, {'name': 'laptop', 'direction': 'front-left', 'distance': 'mid distance', 'orientation': 'facing backward'}, {'name': 'green door', 'direction': 'front-right', 'distance': 'slightly far', 'orientation': 'on front wall'}]
object_positions:
television: [3.0, 2.0]
chair: [3.0, 10.0]
truck: [11.0, 4.0]
bike: [3.0, 6.0]
backpack: [1.0, 12.0]
pan: [9.0, 4.0]
cap: [5.0, 2.0]
lamp: [1.0, 7.0]
vase: [1.0, 14.0]
shelf: [13.0, 2.0]
rubberduck: [5.0, 14.0]
laptop: [9.0, 2.0]
green door: [7.0, 6.0]
blue door: [6.0, 8.0]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
room_id: 3
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: BackwardPovTextEvaluationTask
choices: []
kwargs:
(none)
name_match: True
view_match: True
score: 1
evaluation_mode: use_model_cogmap
Evaluation state
Q2: BackwardPovTextEvaluationTask

Question 3 (ID: 947e7c0517a89154)

✅ Evaluation Results
id: 947e7c0517a89154
question: Now you jump to an object's position, facing its direction. You observe that truck is front, near, facing backward; shelf is front-right, slightly far. Which object are you standing at? Answer format: <object_name> Example: lamp
action: None
answer:
answer: pan
final_pos: [9, 4]
final_ori: [1, 0]
final_observation: [{'name': 'truck', 'direction': 'front', 'distance': 'near', 'orientation': 'facing backward'}, {'name': 'shelf', 'direction': 'front-right', 'distance': 'slightly far', 'orientation': None}]
object_positions:
television: [3.0, 2.0]
chair: [3.0, 10.0]
truck: [11.0, 4.0]
bike: [3.0, 6.0]
backpack: [1.0, 12.0]
pan: [9.0, 4.0]
cap: [5.0, 2.0]
lamp: [1.0, 7.0]
vase: [1.0, 14.0]
shelf: [13.0, 2.0]
rubberduck: [5.0, 14.0]
laptop: [9.0, 2.0]
green door: [7.0, 6.0]
blue door: [6.0, 8.0]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
room_id: 3
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: BackwardPovTextEvaluationTask
choices: []
kwargs:
(none)
name_match: True
view_match: True
score: 1
evaluation_mode: use_model_cogmap
Evaluation state
Q3: BackwardPovTextEvaluationTask

📊 Task: Action2ViewEvaluationTask

Question 1 (ID: 4f8a7d910b60df45)

✅ Evaluation Results
id: 4f8a7d910b60df45
question: You return to your starting position and face north. You will execute the following action sequence: 1. Jump to the object at front-right, mid distance. 2. Rotate(-90) 3. Jump to the object at front-right, mid distance. 4. Rotate(-180) After executing the actions, what is the ego relation of bike relative to you? Answer format: <ego direction>, <distance> Example: front, near
action: None
answer: front-right, mid distance
task_type: Action2ViewEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_model_cogmap
Evaluation state
Q1: Action2ViewEvaluationTask

Question 2 (ID: 32029f3c5e3c31ee)

✅ Evaluation Results
id: 32029f3c5e3c31ee
question: You return to your starting position and face north. You will execute the following action sequence: 1. Rotate(90) 2. Among objects which are front-left, slightly far to you, you jump to the 2nd from left also nearest one. 3. Among objects which are front-right, slightly far to you, you jump to the 2nd from left also farthest one. 4. Rotate(-180) After executing the actions, what is the ego relation of truck relative to you? Answer format: <ego direction>, <distance> Example: front, near
action: None
answer: front-right, mid distance
task_type: Action2ViewEvaluationTask
choices: []
kwargs:
(none)
score: 0
evaluation_mode: use_model_cogmap
Evaluation state
Q2: Action2ViewEvaluationTask

Question 3 (ID: 79787b772d0b8410)

✅ Evaluation Results
id: 79787b772d0b8410
question: You return to your starting position and face north. You will execute the following action sequence: 1. Jump to the object at front-right, mid distance. 2. Rotate(90) 3. Jump to the object at front-left, mid distance. 4. Rotate(90) After executing the actions, what is the ego relation of cap relative to you? Answer format: <ego direction>, <distance> Example: front, near
action: None
answer: front-slight-right, slightly far
task_type: Action2ViewEvaluationTask
choices: []
kwargs:
(none)
score: 0.500
evaluation_mode: use_model_cogmap
Evaluation state
Q3: Action2ViewEvaluationTask

📊 Task: View2ActionTextEvaluationTask

Question 1 (ID: a13a817826736b87)

✅ Evaluation Results
id: a13a817826736b87
question: You return to your starting position and face north. Then you have executed an action sequence and changed to a new location and facing direction. You observe the following: pan is at front-right, slightly far, facing backward; truck is at front-right, mid distance, facing forward; laptop is at front, mid distance, facing backward What action sequence led to this final view? The action sequence must be valid and only contain move actions. Answer format: <sequence of move actions> Example: JumpTo(lamp), Rotate(90)
action: None
answer:
final_pos: [13, 2]
final_ori: [-1, 0]
room_id: 3
init_pos: [1, 3]
init_ori: [0, 1]
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
minimal_plan: [['rotate', 90], ['jumpto', 'green door'], ['jumpto', 'shelf'], ['rotate', 180]]
final_observation: [{'name': 'pan', 'direction': 'front-right', 'distance': 'slightly far', 'orientation': 'facing backward'}, {'name': 'truck', 'direction': 'front-right', 'distance': 'mid distance', 'orientation': 'facing forward'}, {'name': 'laptop', 'direction': 'front', 'distance': 'mid distance', 'orientation': 'facing backward'}]
task_type: View2ActionTextEvaluationTask
choices: []
kwargs:
(none)
pos_match: True
ori_match: True
final_pos: [13.0, 2.0]
final_ori: [-1, 0]
visible_match: True
score: 1
evaluation_mode: use_model_cogmap
Evaluation state
Q1: View2ActionTextEvaluationTask

Question 2 (ID: ef2bc20cb1343cb5)

✅ Evaluation Results
id: ef2bc20cb1343cb5
question: You return to your starting position and face north. Then you have executed an action sequence and changed to a new location and facing direction. You observe the following: blue door is at front-left, mid distance, on front wall What action sequence led to this final view? The action sequence must be valid and only contain move actions. Answer format: <sequence of move actions> Example: JumpTo(lamp), Rotate(90)
action: None
answer:
final_pos: [7, 6]
final_ori: [0, 1]
room_id: [1, 3]
init_pos: [1, 3]
init_ori: [0, 1]
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
minimal_plan: [['rotate', 90], ['jumpto', 'initial_pos'], ['rotate', -90]]
final_observation: [{'name': 'blue door', 'direction': 'front-left', 'distance': 'mid distance', 'orientation': 'on front wall'}]
task_type: View2ActionTextEvaluationTask
choices: []
kwargs:
(none)
pos_match: False
ori_match: True
final_pos: [3.0, 6.0]
final_ori: [0, 1]
visible_match: False
score: 0
evaluation_mode: use_model_cogmap
Evaluation state
Q2: View2ActionTextEvaluationTask

Question 3 (ID: 3d92d2ea666b2ed5)

✅ Evaluation Results
id: 3d92d2ea666b2ed5
question: You return to your starting position and face north. Then you have executed an action sequence and changed to a new location and facing direction. You observe the following: laptop is at front-left, slightly far, facing left; cap is at front-right, slightly far, facing left; pan is at front-left, mid distance, facing left What action sequence led to this final view? The action sequence must be valid and only contain move actions. Answer format: <sequence of move actions> Example: JumpTo(lamp), Rotate(90)
action: None
answer:
final_pos: [7, 6]
final_ori: [0, -1]
room_id: [1, 3]
init_pos: [1, 3]
init_ori: [0, 1]
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
minimal_plan: [['rotate', 90], ['jumpto', 'initial_pos'], ['rotate', 90]]
final_observation: [{'name': 'laptop', 'direction': 'front-left', 'distance': 'slightly far', 'orientation': 'facing left'}, {'name': 'cap', 'direction': 'front-right', 'distance': 'slightly far', 'orientation': 'facing left'}, {'name': 'pan', 'direction': 'front-left', 'distance': 'mid distance', 'orientation': 'facing left'}]
task_type: View2ActionTextEvaluationTask
choices: []
kwargs:
(none)
pos_match: True
ori_match: True
final_pos: [7.0, 6.0]
final_ori: [0, -1]
visible_match: True
score: 1
evaluation_mode: use_model_cogmap
Evaluation state
Q3: View2ActionTextEvaluationTask

📊 Task: AlloMappingEvaluationTask

Question 1 (ID: 47868a3a901e547c)

✅ Evaluation Results
id: 47868a3a901e547c
question: Treat your starting position as the origin (0, 0) while facing north. Report allocentric coordinates using (x right/east, y up/north). Objects: shelf, truck, lamp. Answer format: (x0, y0); (x1, y1); ... in the same order. Example: (1, 0); (-2, 3); (0, -1)
action: None
answer: [[12, -1], [10, 1], [0, 4]]
task_type: AlloMappingEvaluationTask
choices: []
kwargs:
(none)
similarity: 0.715
threshold: 0.900
score: 0.715
evaluation_mode: use_model_cogmap
Evaluation state
Q1: AlloMappingEvaluationTask

Question 2 (ID: adbcc906c717830a)

✅ Evaluation Results
id: adbcc906c717830a
question: Treat your starting position as the origin (0, 0) while facing north. Report allocentric coordinates using (x right/east, y up/north). Objects: rubberduck, shelf, pan. Answer format: (x0, y0); (x1, y1); ... in the same order. Example: (1, 0); (-2, 3); (0, -1)
action: None
answer: [[4, 11], [12, -1], [8, 1]]
task_type: AlloMappingEvaluationTask
choices: []
kwargs:
(none)
similarity: 0.787
threshold: 0.900
score: 0.787
evaluation_mode: use_model_cogmap
Evaluation state
Q2: AlloMappingEvaluationTask

Question 3 (ID: 9cdfd701876daa0f)

✅ Evaluation Results
id: 9cdfd701876daa0f
question: Treat your starting position as the origin (0, 0) while facing north. Report allocentric coordinates using (x right/east, y up/north). Objects: cap, pan, backpack, shelf. Answer format: (x0, y0); (x1, y1); ... in the same order. Example: (1, 0); (-2, 3); (0, -1)
action: None
answer: [[4, -1], [8, 1], [0, 9], [12, -1]]
task_type: AlloMappingEvaluationTask
choices: []
kwargs:
(none)
similarity: 0.577
threshold: 0.900
score: 0.577
evaluation_mode: use_model_cogmap
Evaluation state
Q3: AlloMappingEvaluationTask

📊 Task: RotEvaluationTask

Question 1 (ID: c7b0da8750e06c67)

✅ Evaluation Results
id: c7b0da8750e06c67
question: You return to your starting position and face north. You will perform a full 360-degree rotation by continuously turning counterclockwise in place. Assume all walls are removed (you can see through walls), so every object is visible. Focus on this set of objects: bike, pan, television. List them in the exact order they appear directly ahead while you rotate. If two objects share a bearing, place the nearer one first. Answer format: <object_name1>, <object_name2>, ... Example: mug, sofa, plant
action: None
answer: ['television', 'pan', 'bike']
task_type: RotEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_model_cogmap
Evaluation state
Q1: RotEvaluationTask

Question 2 (ID: fc92fab04fedd039)

✅ Evaluation Results
id: fc92fab04fedd039
question: You return to your starting position and face north. You will perform a full 360-degree rotation by continuously turning counterclockwise in place. Assume all walls are removed (you can see through walls), so every object is visible. Focus on this set of objects: bike, television, truck. List them in the exact order they appear directly ahead while you rotate. If two objects share a bearing, place the nearer one first. Answer format: <object_name1>, <object_name2>, ... Example: mug, sofa, plant
action: None
answer: ['television', 'truck', 'bike']
task_type: RotEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_model_cogmap
Evaluation state
Q2: RotEvaluationTask

Question 3 (ID: 83da2dae108d2082)

✅ Evaluation Results
id: 83da2dae108d2082
question: You return to your starting position and face north. You will perform a full 360-degree rotation by continuously turning counterclockwise in place. Assume all walls are removed (you can see through walls), so every object is visible. Focus on this set of objects: chair, television, truck. List them in the exact order they appear directly ahead while you rotate. If two objects share a bearing, place the nearer one first. Answer format: <object_name1>, <object_name2>, ... Example: mug, sofa, plant
action: None
answer: ['television', 'truck', 'chair']
task_type: RotEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_model_cogmap
Evaluation state
Q3: RotEvaluationTask

📊 Task: Location2ViewEvaluationTask

Question 1 (ID: 13d7589043fe1ee0)

✅ Evaluation Results
id: 13d7589043fe1ee0
question: Treat the green door as the new 'origin' (0, 0). You move to (2, -5) and face north. What is the egocentric relation of pan? Answer format: <direction>, <distance> Example: front, near
action: None
answer: front, mid distance
task_type: Location2ViewEvaluationTask
choices: []
kwargs:
(none)
score: 0.500
evaluation_mode: use_model_cogmap
Evaluation state
Q1: Location2ViewEvaluationTask

Question 2 (ID: e2c7ab6a2049c06a)

✅ Evaluation Results
id: e2c7ab6a2049c06a
question: Still treat your initial position as origin (0, 0) You move to (4, 4) and face west. What is the egocentric relation of bike? Answer format: <direction>, <distance> Example: front, near
action: None
answer: front-left, mid distance
task_type: Location2ViewEvaluationTask
choices: []
kwargs:
(none)
score: 1
evaluation_mode: use_model_cogmap
Evaluation state
Q2: Location2ViewEvaluationTask

Question 3 (ID: 5a56a83a64962b71)

✅ Evaluation Results
id: 5a56a83a64962b71
question: Still treat your initial position as origin (0, 0) You move to (0, 3) and face east. What is the egocentric relation of blue door? Answer format: <direction>, <distance> Example: front, near
action: None
answer: front-slight-left, slightly far
task_type: Location2ViewEvaluationTask
choices: []
kwargs:
(none)
score: 0
evaluation_mode: use_model_cogmap
Evaluation state
Q3: Location2ViewEvaluationTask

📊 Task: View2LocationTextEvaluationTask

Question 1 (ID: 19a422deab9e39fc)

✅ Evaluation Results
id: 19a422deab9e39fc
question: You move to a new location and face north. You observe: pan is at front, mid distance, facing right; truck is at front-right, mid distance, facing left; green door is at front-slight-left, slightly far, on left wall Treat the green door as the new 'origin' (0, 0). What is your new 2D coordinate (x, y)? Answer format: (x, y) Example: (2, -1)
action: None
answer:
coord: [2, -5]
final_pos: [9, 1]
final_ori: [0, 1]
room_id: 3
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
final_observation: [{'name': 'pan', 'direction': 'front', 'distance': 'mid distance', 'orientation': 'facing right'}, {'name': 'truck', 'direction': 'front-right', 'distance': 'mid distance', 'orientation': 'facing left'}, {'name': 'green door', 'direction': 'front-slight-left', 'distance': 'slightly far', 'orientation': 'on left wall'}]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: View2LocationTextEvaluationTask
choices: []
kwargs:
(none)
similarity: 0.831
raw_score: 0.831
score: 0.831
evaluation_mode: use_model_cogmap
Evaluation state
Q1: View2LocationTextEvaluationTask

Question 2 (ID: 197270037eaa3b24)

✅ Evaluation Results
id: 197270037eaa3b24
question: You move to a new location and face west. You observe: bike is at front-left, mid distance, facing forward; lamp is at front, mid distance Still treat your initial position as origin (0, 0) What is your new 2D coordinate (x, y)? Answer format: (x, y) Example: (2, -1)
action: None
answer:
coord: [4, 4]
final_pos: [5, 7]
final_ori: [-1, 0]
room_id: 1
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
final_observation: [{'name': 'bike', 'direction': 'front-left', 'distance': 'mid distance', 'orientation': 'facing forward'}, {'name': 'lamp', 'direction': 'front', 'distance': 'mid distance', 'orientation': None}]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: View2LocationTextEvaluationTask
choices: []
kwargs:
(none)
similarity: 0.838
raw_score: 0.838
score: 0.838
evaluation_mode: use_model_cogmap
Evaluation state
Q2: View2LocationTextEvaluationTask

Question 3 (ID: 7d2a51a882153983)

✅ Evaluation Results
id: 7d2a51a882153983
question: You move to a new location and face east. You observe: blue door is at front-slight-left, slightly far, on left wall; cap is at front-right, slightly far, facing forward; bike is at front, near, facing backward Still treat your initial position as origin (0, 0) What is your new 2D coordinate (x, y)? Answer format: (x, y) Example: (2, -1)
action: None
answer:
coord: [0, 3]
final_pos: [1, 6]
final_ori: [1, 0]
room_id: 1
object_positions:
television: [3, 2]
chair: [3, 10]
truck: [11, 4]
bike: [3, 6]
backpack: [1, 12]
pan: [9, 4]
cap: [5, 2]
lamp: [1, 7]
vase: [1, 14]
shelf: [13, 2]
rubberduck: [5, 14]
laptop: [9, 2]
green door: [7, 6]
blue door: [6, 8]
object_orientations:
television: [1, 0]
chair: [1, 0]
truck: [-1, 0]
bike: [-1, 0]
backpack: [0, -1]
pan: [1, 0]
cap: [1, 0]
rubberduck: [-1, 0]
laptop: [1, 0]
green door: [1, 0]
blue door: [0, 1]
final_observation: [{'name': 'blue door', 'direction': 'front-slight-left', 'distance': 'slightly far', 'orientation': 'on left wall'}, {'name': 'cap', 'direction': 'front-right', 'distance': 'slightly far', 'orientation': 'facing forward'}, {'name': 'bike', 'direction': 'front', 'distance': 'near', 'orientation': 'facing backward'}]
gate_info:
green door:
room_ids: [1, 3]
ori_by_room:
1: [-1, 0]
3: [1, 0]
blue door:
room_ids: [1, 2]
ori_by_room:
1: [0, -1]
2: [0, 1]
object_rooms:
television: 1
chair: 2
truck: 3
bike: 1
backpack: 2
pan: 3
cap: 1
lamp: 1
vase: 2
shelf: 3
rubberduck: 2
laptop: 3
green door: [1, 3]
blue door: [1, 2]
task_type: View2LocationTextEvaluationTask
choices: []
kwargs:
(none)
similarity: 0.717
raw_score: 0.717
score: 0.717
evaluation_mode: use_model_cogmap
Evaluation state
Q3: View2LocationTextEvaluationTask