Locomotion on D4RL halfcheetah-medium-replay
0.8874Normalized ScoreOff2On
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Off2OnOnline update way=SAC2024.12 | 0.8874 | 0.4318 | — | |
| O2SACOnline update way=SAC2024.12 | 0.7165 | 0.2637 | — | |
| QDQGradient steps=1M2026.05 | 0.637 | — | — | |
| EDACGradient steps=3M2026.05 | 0.613 | — | — | |
| CPQLGradient steps=1M2026.05 | 0.603 | — | — | |
| PRDC2024.03 | 0.55 | — | — | |
| MOPO2026.03 | 0.531 | — | — | |
| OFQLCategory=One-step Flow Policies2025.08 | 0.512 | — | — | |
| FQLCategory=One-step Flow Policies2025.08 | 0.511 | — | — | |
| PBRLGradient steps=1M2026.05 | 0.495 | — | — | |
| DQLCategory=Multi-step Diffusion Policies2025.08 | 0.478 | — | — | |
| O2PPOOnline update way=IQL2024.12 | 0.4676 | 0.0422 | — | |
| IDQLCategory=Multi-step Diffusion Policies2025.08 | 0.465 | — | — | |
| SAGE2026.03 | 0.465 | — | — | |
| DMEMM2026.02 | 0.461 | — | — | |
| DVreproduced=true2026.03 | 0.458 | — | — | |
| CQL2026.02 | 0.455 | — | — | |
| CQL2026.02 | 0.455 | — | — | |
| CQL2026.03 | 0.455 | — | — | |
| EDPCategory=Multi-step Diffusion Policies2025.08 | 0.449 | — | — | |
| HDMI2026.02 | 0.449 | — | — | |
| A2PO2024.03 | 0.448 | — | — | |
| TD3-BCCategory=Non-Diffusion Policies2025.08 | 0.446 | — | — | |
| IQL+ReD2024.03 | 0.443 | — | — | |
| IQL2026.02 | 0.442 | — | — | |
| IQLCategory=Non-Diffusion Policies2025.08 | 0.442 | — | — | |
| IQL2026.02 | 0.442 | — | — | |
| IQL2026.03 | 0.442 | — | — | |
| TD3+BC2026.02 | 0.433 | — | — | |
| DDCategory=Diffusion Planners2025.08 | 0.429 | — | — | |
| MBOP2026.03 | 0.423 | — | — | |
| TT2026.02 | 0.419 | — | — | |
| TT2026.03 | 0.419 | — | — | |
| TTGradient steps=1M2026.05 | 0.419 | — | — | |
| LDCQ2026.03 | 0.418 | — | — | |
| ISCTSeeds=5, Episodes=102026.02 | 0.41 | — | — | |
| RGG2026.03 | 0.41 | — | — | |
| %BC2024.03 | 0.406 | — | — | |
| MOReL2026.02 | 0.402 | — | — | |
| MOReL2026.03 | 0.402 | — | — | |
| IMLE+Exp RW2026.03 | 0.395 | — | — | |
| LoMAP2026.03 | 0.391 | — | — | |
| HD-DA2026.02 | 0.381 | — | — | |
| Diffuser2026.02 | 0.377 | — | — | |
| Diffuserreproduced=true2026.03 | 0.377 | — | — | |
| Diffuser2026.03 | 0.377 | — | — | |
| DT2024.03 | 0.366 | — | — | |
| DT2026.02 | 0.366 | — | — | |
| BCCategory=Non-Diffusion Policies2025.08 | 0.366 | — | — | |
| BC2026.02 | 0.366 | — | — | |
| DT2026.02 | 0.366 | — | — | |
| BC2026.03 | 0.366 | — | — | |
| DT2026.03 | 0.366 | — | — | |
| DTGradient steps=1M2026.05 | 0.366 | — | — | |
| DiffuserCategory=Diffusion Planners2025.08 | 0.36 | — | — | |
| UWACGradient steps=1M2026.05 | 0.359 | — | — | |
| Tgt+Src (Edited)Domain Gap=Thigh Size2024.09 | 0.33 | 26 | — | |
| Tgt+Src (Edited)Domain Gap=Friction2024.09 | 0.318 | 21.4 | — | |
| Tgt+Src (Edited)Domain Gap=Gravity2024.09 | 0.313 | 19.5 | — | |
| Tgt+SrcDomain Gap=Gravity2024.09 | 0.295 | 12.6 | — | |
| Tgt+SrcDomain Gap=Thigh Size2024.09 | 0.29 | 10.7 | — | |
| TgtDomain Gap=Gravity2024.09 | 0.262 | — | — | |
| TgtDomain Gap=Friction2024.09 | 0.262 | — | — | |
| TgtDomain Gap=Thigh Size2024.09 | 0.262 | — | — | |
| Tgt+SrcDomain Gap=Friction2024.09 | 0.232 | -11.5 | — | |
| SrcDomain Gap=Friction2024.09 | 0.144 | — | — | |
| SrcDomain Gap=Gravity2024.09 | 0.084 | — | — | |
| SrcDomain Gap=Thigh Size2024.09 | -0.001 | — | — | |
| AWAC2026.05 | — | — | 2,189.6 | |
| BaseRL Algorithm=CQL2025.06 | — | — | 45.9 | |
| BaseRL Algorithm=DT2025.06 | — | — | 38.8 | |
| BC2026.05 | — | — | 1,967.7 | |
| BiTrajDiffRL Algorithm=CQL2025.06 | — | — | 45.8 | |
| BiTrajDiffRL Algorithm=DT2025.06 | — | — | 39.5 | |
| Cal-QL2026.05 | — | — | 1,795.2 | |
| COOPO2026.05 | — | — | 4,602.5 | |
| CQL2026.05 | — | — | 2,477.9 | |
| DiffStitchRL Algorithm=CQL2025.06 | — | — | 45.1 | |
| DiffStitchRL Algorithm=DT2025.06 | — | — | 38.9 | |
| IQL2026.05 | — | — | 3,368.4 | |
| ODT2026.05 | — | — | 2,087.6 | |
| Off2ON2026.05 | — | — | 2,747.4 | |
| Onestep RL2026.05 | — | — | 2,053.5 | |
| PEX2026.05 | — | — | 2,995.8 | |
| RTDiffRL Algorithm=CQL2025.06 | — | — | 45.6 | |
| RTDiffRL Algorithm=DT2025.06 | — | — | 39.1 | |
| SyntherRL Algorithm=CQL2025.06 | — | — | 47.4 | |
| SyntherRL Algorithm=DT2025.06 | — | — | 39.3 | |
| TD3+BC2026.05 | — | — | 2,423.5 | |
| Uni-O42026.05 | — | — | 2,911.2 |