| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Open door | Meta-World | VOC Score62.64 | 35 | |
| Reward Modeling | Meta-World Open door | Prediction Accuracy65.46 | 28 | |
| Reward Modeling | Meta-World Open drawer | Prediction Accuracy69.01 | 28 | |
| Reward Modeling | Meta-World Button press | Prediction Accuracy76.44 | 28 | |
| Open drawer | Meta-World | VOC Score90.17 | 28 | |
| Button press | Meta-World | VOC Score95.47 | 28 | |
| Robotic Manipulation | Meta-World | Average Success Rate94.7 | 27 | |
| Robot Manipulation | Meta-World | Latency (Easy) (ms)10.1 | 15 | |
| Multi-task Reinforcement Learning | Meta-World MT10 v1 (Fixed) | Success Rate88 | 12 | |
| Offline Reinforcement Learning | Meta-World medium-replay | BP->DC*3,967 | 10 | |
| Multi-Task Reinforcement Learning | Meta-World MT10 v1 (train test) | Average Success91 | 9 | |
| reach | Meta-World ML-1 (test) | Success Rate100 | 9 | |
| Multi-task Reinforcement Learning | Meta-World MT50 (MT50-rand) V2 (Near-optimal) | Avg Success Rate61.32 | 8 | |
| Task Generalization | Meta-World ML-45 (test) | Success Rate81.7 | 8 | |
| Task Generalization | Meta-World ML-10 (test) | Success Rate97.5 | 8 | |
| Multi-task Reinforcement Learning | Meta-World MT50 v1 (Fixed) | Success Rate60 | 8 | |
| door-unlock | Meta-World v2 (test) | Best Attack Reward3,421 | 7 | |
| door-lock | Meta-World v2 (test) | Best Attack Reward2,043 | 7 | |
| handle-pull-side | Meta-World v2 (test) | Best Attack Reward4,268 | 7 | |
| handle-press-side | Meta-World v2 (test) | Best Attack Reward4,726 | 7 | |
| faucet-open | Meta-World v2 (test) | Best Attack Reward4,383 | 7 | |
| faucet-close | Meta-World v2 (test) | Best Attack Reward4,108 | 7 | |
| drawer-open | Meta-World v2 (test) | Best Attack Reward1,556 | 7 | |
| drawer-close | Meta-World v2 (test) | Best Attack Reward4,868 | 7 | |
| window-open | Meta-World v2 (test) | Best Attack Reward671 | 7 |