Dexterous Manipulation on Adroit Pen
98Success RateWSRL
Evaluation Results
| Method | Links | |
|---|---|---|
| WSRLData Availability=Without Data, Training Steps=300k2026.05 | 98 | |
| IBRLData Availability=Without Data, Training Steps=300k2026.05 | 95 | |
| CalQLData Availability=Without Data, Training Steps=300k2026.05 | 93 | |
| Q2RLData Availability=Without Data, Training Steps=300k2026.05 | 93 | |
| WSRLData Availability=Without Data, Training Steps=150k2026.05 | 92 | |
| Q2RLData Availability=Without Data, Training Steps=150k2026.05 | 91 | |
| BC PolicyData Availability=Without Data, Training Steps=02026.05 | 90 | |
| BC PolicyData Availability=Without Data, Training Steps=150k2026.05 | 90 | |
| BC PolicyData Availability=Without Data, Training Steps=300k2026.05 | 90 | |
| Q2RLData Availability=Without Data, Training Steps=02026.05 | 88 | |
| CalQLData Availability=Without Data, Training Steps=150k2026.05 | 87 | |
| IBRLData Availability=Without Data, Training Steps=150k2026.05 | 82 | |
| WSRLData Availability=Without Data, Training Steps=02026.05 | 78 | |
| DP3-VADFTraining Steps=20,0002026.04 | 76.7 | |
| CalQLData Availability=Without Data, Training Steps=02026.05 | 71 | |
| CQLData Availability=Without Data, Training Steps=02026.05 | 70 | |
| DP3Training Steps=20,0002026.04 | 65.8 | |
| Dexora2026.05 | 65 | |
| GR00T N12026.05 | 45 | |
| π02026.05 | 20 | |
| CQLData Availability=Without Data, Training Steps=150k2026.05 | 12 | |
| IBRLData Availability=Without Data, Training Steps=02026.05 | 10 | |
| IBCTraining Steps=20,0002026.04 | 9 | |
| BCRNNTraining Steps=20,0002026.04 | 9 | |
| CQLData Availability=Without Data, Training Steps=300k2026.05 | 5 | |
| DP2026.05 | 5 |