Robotic Manipulation on Cube (test)
79Success RatePRISM-MPPI (ours)
Evaluation results
| Method | Links | |
|---|---|---|
| PRISM-MPPI (ours)Learned-prior params=1.0M, Sample budget (K)=322026.06 | 79 | |
| PRISM-MPPI (ours)Learned-prior params=1.0M, Sample budget (K)=1282026.06 | 79 | |
| PRISM-MPPI (ours)Learned-prior params=1.0M, Sample budget (K)=642026.06 | 78 | |
| Diffusion PolicyLearned-prior params=19.3M, Sample budget (K)=K-independent2026.06 | 77 | |
| PRISM-MPPI (DINO-WM-style)Learned-prior params=1.0M, Sample budget (K)=1282026.06 | 67 | |
| BC-only (head µ alone)Learned-prior params=1.0M, Sample budget (K)=K-independent2026.06 | 66 | |
| PRISM-MPPI (DINO-WM-style)Learned-prior params=1.0M, Sample budget (K)=642026.06 | 65 | |
| PRISM-MPPI (DINO-WM-style)Learned-prior params=1.0M, Sample budget (K)=322026.06 | 57 | |
| Vanilla MPPI (DINO-WM-style)Learned-prior params=0, Sample budget (K)=642026.06 | 49 | |
| Vanilla MPPI (LeWM)Learned-prior params=0, Sample budget (K)=322026.06 | 46 | |
| Vanilla MPPI (DINO-WM-style)Learned-prior params=0, Sample budget (K)=322026.06 | 45 | |
| Vanilla MPPI (LeWM)Learned-prior params=0, Sample budget (K)=642026.06 | 44 | |
| Vanilla MPPI (LeWM)Learned-prior params=0, Sample budget (K)=1282026.06 | 44 | |
| Vanilla MPPI (DINO-WM-style)Learned-prior params=0, Sample budget (K)=1282026.06 | 41 |