Value Estimation on Suboptimal-Value-Bench ARX
0.05Hesitation RMSEWVM
Evaluation Results
| Method | Links | |
|---|---|---|
| WVMEmbodiment=ARX2026.06 | 0.05 | |
| RobometerEmbodiment=ARX2026.06 | 0.12 | |
| GVLEmbodiment=ARX2026.06 | 0.14 | |
| RoboRewardEmbodiment=ARX2026.06 | 0.17 | |
| TopRewardEmbodiment=ARX2026.06 | 0.24 | |
| VLACEmbodiment=ARX2026.06 | 0.5 | |
| Robo-DopamineEmbodiment=ARX2026.06 | 0.52 |