Share your thoughts, 1 month free Claude Pro on usSee more

SwimmerVelocity on Safety Gym

0.94Normalized Reward

Reference policy (πref)

Updated 13d ago

Evaluation Results

Method	Links
Reference policy (πref) 2026.05		0.94	6.34
PREFINE 2026.05		0.62	0.36
BC 2026.05		0.52	4.83
SafeDICE 2026.05		0.5	0.2
CPQ 2026.05		0.14	2.79
PPL 2026.05		-0.03	1.25