List your company
ResearchFormerly WizWand

Navigation on D4RL antmaze-medium-play

88.1Normalized ScoreOFQL
-3.52420.26344.0567.837Oct 15, 2022Apr 6, 2023Sep 26, 2023Mar 17, 2024Sep 6, 2024Feb 26, 2025Aug 19, 2025
Updated 8mo ago

Evaluation results

MethodLinks
OFQLCategory=One-step Flow Policies2025.0888.1
PORMethod Category=Conditioned BC, Seeds=52022.1084.6
FQLCategory=One-step Flow Policies2025.0878
C-LAPImplementation=Reproduced2024.1177.5
DQLCategory=Multi-step Diffusion Policies2025.0876.6
EDPCategory=Multi-step Diffusion Policies2025.0873.3
IQLMethod Category=Weighted BC, Seeds=52022.1071.2
IQLCategory=Non-Diffusion Policies2025.0871.2
CQLMethod Category=Best RL Baseline, Seeds=52022.1068
IDQLCategory=Multi-step Diffusion Policies2025.0867.3
RvSMethod Category=Conditioned BC, Seeds=52022.1058.1
TD3-BCCategory=Non-Diffusion Policies2025.0810.6
DDCategory=Diffusion Planners2025.088
DiffuserCategory=Diffusion Planners2025.086.7
10%BCMethod Category=Weighted BC, Seeds=52022.105.4
One-step RLMethod Category=Weighted BC, Seeds=52022.100.3
PLASImplementation=Original Paper2024.110.2
DTMethod Category=Conditioned BC, Seeds=52022.100
PLASImplementation=Reproduced2024.110
MOPOImplementation=Reproduced2024.110
MOBILEImplementation=Reproduced2024.110
BCCategory=Non-Diffusion Policies2025.080