List your company

PointPush1 on Safety-Gymnasium

49Normalized RewardFISOR
-0.9212.042537.96Apr 29, 2026

Evaluation results

MethodLinks
FISOR2026.044983
BCQ-Lag2026.043386
BCQ-Lagcost thresholds=3, evaluation episodes=20, random seeds=32026.043386
DT + SASBackbone=DT, Alignment strategy (SAS)=Lyapunov-Guided2026.042862
DT + SAScost thresholds=3, evaluation episodes=20, random seeds=32026.042862
CDT+SASAlignment=SAS, Base Model=CDT2026.042654
CDT + SASBackbone=CDT, Alignment strategy (SAS)=Lyapunov-Guided2026.042654
CDT + SAScost thresholds=3, evaluation episodes=20, random seeds=32026.042654
CDTBackbone=CDT2026.042448
CDTcost thresholds=3, evaluation episodes=20, random seeds=32026.042448
BEAR-Lag2026.042279
BEAR-Lagcost thresholds=3, evaluation episodes=20, random seeds=32026.042279
CPQ2026.042083
CPQcost thresholds=3, evaluation episodes=20, random seeds=32026.042083
BC-All2026.041961
BC-Allcost thresholds=3, evaluation episodes=20, random seeds=32026.041961
BC-Safe2026.041343
COptiDICE2026.041383
BC-Safecost thresholds=3, evaluation episodes=20, random seeds=32026.041343
COptiDICEcost thresholds=3, evaluation episodes=20, random seeds=32026.041383
DCRL2026.04152