List your company

PointButton2 on Safety-Gymnasium

58Normalized RewardCPQ
-1.2814.1129.544.89Apr 29, 2026

Evaluation results

MethodLinks
CPQ2026.04584.3
CPQcost thresholds=3, evaluation episodes=20, random seeds=32026.04584.3
DT + SASBackbone=DT, Alignment strategy (SAS)=Lyapunov-Guided2026.04511.14
DT + SAScost thresholds=3, evaluation episodes=20, random seeds=32026.04511.14
CDTBackbone=CDT2026.04461.57
CDTcost thresholds=3, evaluation episodes=20, random seeds=32026.04461.57
BEAR-Lag2026.04432.47
BEAR-Lagcost thresholds=3, evaluation episodes=20, random seeds=32026.04432.47
CDT+SASAlignment=SAS, Base Model=CDT2026.044198
CDT + SASBackbone=CDT, Alignment strategy (SAS)=Lyapunov-Guided2026.04410.98
CDT + SAScost thresholds=3, evaluation episodes=20, random seeds=32026.04410.98
BCQ-Lag2026.04402.66
BCQ-Lagcost thresholds=3, evaluation episodes=20, random seeds=32026.04402.66
BC-All2026.04272.02
BC-Allcost thresholds=3, evaluation episodes=20, random seeds=32026.04272.02
DCRL2026.04180.64
BC-Safe2026.04161.1
BC-Safecost thresholds=3, evaluation episodes=20, random seeds=32026.04161.1
COptiDICE2026.04151.51
COptiDICEcost thresholds=3, evaluation episodes=20, random seeds=32026.04151.51
FISOR2026.04158