List your company

PointGoal1 on Safety-Gymnasium

0.74Normalized RewardBEAR-Lag
0.220.3550.490.625Apr 29, 2026

Evaluation results

MethodLinks
BEAR-LagCost thresholds=3, Evaluation episodes=20, Random seeds=32026.040.741.18
BEAR-Lagcost thresholds=3, evaluation episodes=20, random seeds=32026.040.741.18
BCQ-LagCost thresholds=3, Evaluation episodes=20, Random seeds=32026.040.710.98
BCQ-Lagcost thresholds=3, evaluation episodes=20, random seeds=32026.040.710.98
CDTBackbone=CDT, Alignment strategy (SAS)=None, Cost thresholds=3, Evaluation episodes=20, Random seeds=32026.040.691.12
CDTcost thresholds=3, evaluation episodes=20, random seeds=32026.040.691.12
DT + SASBackbone=DT, Alignment strategy (SAS)=Lyapunov-Guided, Cost thresholds=3, Evaluation episodes=20, Random seeds=32026.040.661.19
DT + SAScost thresholds=3, evaluation episodes=20, random seeds=32026.040.661.19
CDT+SASAlignment=SAS, Base Model=CDT2026.040.651.27
CDT + SASBackbone=CDT, Alignment strategy (SAS)=Lyapunov-Guided, Cost thresholds=3, Evaluation episodes=20, Random seeds=32026.040.651.27
BC-AllCost thresholds=3, Evaluation episodes=20, Random seeds=32026.040.650.95
CDT + SAScost thresholds=3, evaluation episodes=20, random seeds=32026.040.651.27
BC-Allcost thresholds=3, evaluation episodes=20, random seeds=32026.040.650.95
CPQCost thresholds=3, Evaluation episodes=20, Random seeds=32026.040.570.35
CPQcost thresholds=3, evaluation episodes=20, random seeds=32026.040.570.35
COptiDICECost thresholds=3, Evaluation episodes=20, Random seeds=32026.040.491.66
COptiDICEcost thresholds=3, evaluation episodes=20, random seeds=32026.040.491.66
BC-SafeCost thresholds=3, Evaluation episodes=20, Random seeds=32026.040.430.54
BC-Safecost thresholds=3, evaluation episodes=20, random seeds=32026.040.430.54
FISOR2026.040.30.35
DCRLCost thresholds=3, Evaluation episodes=20, Random seeds=32026.040.240.86