State Probing on RoboSuite Stacking
0.78R2 ScoreMIST-WM
Evaluation Results
| Method | Links | |
|---|---|---|
| MIST-WMEvaluation Protocol=Linear Probing2026.07 | 0.78 | |
| METRAEvaluation Protocol=Linear Probing2026.07 | 0.59 | |
| Denoised MDPEvaluation Protocol=Linear Probing2026.07 | 0.56 | |
| I-FactorEvaluation Protocol=Linear Probing2026.07 | 0.48 | |
| Plan2ExploreEvaluation Protocol=Linear Probing2026.07 | 0.45 | |
| Curiosity ReplayEvaluation Protocol=Linear Probing2026.07 | 0.44 | |
| DINO-WMEvaluation Protocol=Linear Probing2026.07 | 0.4 | |
| RandomEvaluation Protocol=Linear Probing2026.07 | 0.31 |