Locomotion on D4RL halfcheetah-medium-expert (Test Return)
9,242.2Test ReturnCOOPO
Evaluation Results
| Method | Links | |
|---|---|---|
| COOPO2026.05 | 9,242.2 | |
| Onestep RL2026.05 | 8,949.3 | |
| Uni-O42026.05 | 8,859.3 | |
| Cal-QL2026.05 | 8,782.9 | |
| CQL2026.05 | 8,762.8 | |
| TD3+BC2026.05 | 8,674.5 | |
| IQL2026.05 | 8,278.3 | |
| BC2026.05 | 5,192.7 | |
| Off2ON2026.05 | 4,333 | |
| AWAC2026.05 | 3,976.5 | |
| ODT2026.05 | 3,566.1 | |
| PEX2026.05 | 2,778.3 | |
| BiTrajDiffRL Algorithm=CQL2025.06 | 94.1 | |
| SyntherRL Algorithm=CQL2025.06 | 92.2 | |
| DiffStitchRL Algorithm=CQL2025.06 | 91.4 | |
| RTDiffRL Algorithm=CQL2025.06 | 90 | |
| BaseRL Algorithm=CQL2025.06 | 87.2 | |
| BiTrajDiffRL Algorithm=DT2025.06 | 83.5 | |
| SyntherRL Algorithm=DT2025.06 | 80.4 | |
| DiffStitchRL Algorithm=DT2025.06 | 76.9 | |
| RTDiffRL Algorithm=DT2025.06 | 72.3 | |
| BaseRL Algorithm=DT2025.06 | 70.9 |