Scenario
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Scenario 1
1.017Policy Value (PV)
15
Scenario 5
90ASR
12
Scenario 3 (S3)
1ASR
12
Scenario 2 (S2)
1ASR
12
Scenario 1 (S1)
100ASR
12
10 agent scenario (ground truth goals)
2.31Trajectory Success Rate
12
Scenario 1 T=300
1Prop. K_hat != K
10
Scenario 1 T=150
0Error Proportion
10
Scenario S4
100Success Rate
9
Scenario S2
100Success Rate
9
Scenario S1
100Success Rate
9
Scenario B (nonlinear, SVM)
1.7False Claim Rate (FCR)
8
Scenario 1
0.168MAE (τ=0.05)
8
Scenario 2 Model A
0.038MSE (τ=0.05)
8
Scenario 1 Model A
0.031MSE (τ=0.05)
8
Scenario A Out-of-Distribution 500 g
0.301RMSE (Slow 0.5 m/s)
8
Scenario A In-Distribution 300 g
0.215RMSE (Slow 0.5 m/s)
8
Scenario S3 Crisis Shock
8.18Average Reward
8
Scenario 3 (two Franka Panda manipulators) 1.0 (test)
100Success Rate
8
Scenario 2 (Two Franka Panda manipulators with closed-chain constraints) (test)
100Success Rate
8
Scenario 4
0.926Similarity (95th Percentile)
8
Scenario VISSIM corridor 1
1,749.57ANP
7
Scenario C
98.7Finish Rate Avg
7
Scenario III
100Success Rate
6
Scenario II
100Success Rate
6