List your company
ResearchFormerly WizWand

DroneRun on Bullet Gym

0.84Normalized RewardReference policy (πref)
0.10160.29330.4850.6767May 20, 2026
Updated 5mo ago

Evaluation results

MethodLinks
Reference policy (πref)2026.050.846.77
PREFINE2026.050.570.42
BC2026.050.550.01
PPL2026.050.540.79
SafeDICE2026.050.460.4
CPQ2026.050.130.04