List your company

PointButton1 on Safety-Gymnasium

69Normalized RewardCPQ
-1.7216.643553.36Apr 29, 2026

Evaluation results

MethodLinks
CPQ2026.04693.2
CPQcost thresholds=3, evaluation episodes=20, random seeds=32026.04693.2
CDT+SASAlignment=SAS, Base Model=CDT2026.04511.27
CDT + SASBackbone=CDT, Alignment strategy (SAS)=Lyapunov-Guided2026.04511.27
CDT + SAScost thresholds=3, evaluation episodes=20, random seeds=32026.04511.27
CDTBackbone=CDT2026.04501.68
CDTcost thresholds=3, evaluation episodes=20, random seeds=32026.04501.68
DT + SASBackbone=DT, Alignment strategy (SAS)=Lyapunov-Guided2026.04491.38
DT + SAScost thresholds=3, evaluation episodes=20, random seeds=32026.04491.38
BCQ-Lag2026.04241.73
BCQ-Lagcost thresholds=3, evaluation episodes=20, random seeds=32026.04241.73
BEAR-Lag2026.04201.6
BEAR-Lagcost thresholds=3, evaluation episodes=20, random seeds=32026.04201.6
COptiDICE2026.04131.4
COptiDICEcost thresholds=3, evaluation episodes=20, random seeds=32026.04131.4
BC-All2026.041010.5
BC-Allcost thresholds=3, evaluation episodes=20, random seeds=32026.041010.5
BC-Safe2026.0460.52
BC-Safecost thresholds=3, evaluation episodes=20, random seeds=32026.0460.52
FISOR2026.0410.58
DCRL2026.0410.48