Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Spatial and Temporal Simulation Tasks

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multi-task EvaluationSpatial and Temporal Simulation Tasks Episode-level
Average Total Score7.81
7
Counting TaskSpatial and Temporal Simulation Tasks Episode-level
Subtask Accuracy4.203
7
Long-horizon TaskSpatial and Temporal Simulation Tasks Episode-level
Subtask Accuracy3.576
7
Spatial InformationSpatial and Temporal Simulation Tasks Episode-level
Subtask Accuracy3.543
7
Multi-task EvaluationSpatial and Temporal Simulation Tasks Step-level
Average Total Score8.318
7
Counting TaskSpatial and Temporal Simulation Tasks Step-level
Subtask Accuracy4.237
7
Long-horizon TaskSpatial and Temporal Simulation Tasks Step-level
Subtask Accuracy3.909
7
Spatial InformationSpatial and Temporal Simulation Tasks Step-level
Subtask Accuracy3.8
7
Showing 8 of 8 rows