Physical Reasoning on PHYRE-1B cross-template (test)
42.2AUCCESSRPIN
Evaluation Results
| Method | Links | |
|---|---|---|
| RPIN2021.02 | 42.2 | |
| Dynamics-Aware DQNLearning Mode=Self-supervised2021.02 | 41.9 | |
| Dec [Joint]2021.02 | 40.3 | |
| Dynamics-Aware DQNLearning Mode=Hand-crafted2021.02 | 38.6 | |
| DQN2021.02 | 36.8 | |
| MEM2021.02 | 18.5 | |
| RAND2021.02 | 13 |