Robotic Manipulation on Robomimic Can (mh) 2D state-based
98Success Rate (Max Perf)DP
Evaluation results
| Method | Links | ||
|---|---|---|---|
| DPNFE=100, Observation Modality=2D state-based, Network Architecture=Transformer, CNN (average)2026.05 | 98 | 92 | |
| IDPNFE=1, Observation Modality=2D state-based, Network Architecture=Transformer, CNN (average)2026.05 | 98 | 94 | |
| FlowNFE=9, Observation Modality=2D state-based, Network Architecture=Transformer, CNN (average)2026.05 | 96 | 92 | |
| MIPNFE=2, Observation Modality=2D state-based, Network Architecture=Transformer, CNN (average)2026.05 | 96 | 94 | |
| MP1NFE=1, Observation Modality=2D state-based, Network Architecture=Transformer, CNN (average)2026.05 | 96 | 87 | |
| Naive DriftingNFE=1, Observation Modality=2D state-based, Network Architecture=Transformer, CNN (average)2026.05 | 96 | 91 | |
| CPNFE=1, Observation Modality=2D state-based, Network Architecture=Transformer, CNN (average)2026.05 | 83 | 74 |