Visual Reasoning On Phyre 1B Cross
평가 지표
AUCCESS
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | AUCCESS | Paper Title | Repository |
|---|---|---|---|
| RPIN | 42.2 | Learning Long-term Visual Dynamics with Region Proposal Interaction Networks | - |
| Dec[Joint]1f | 40.3 | Forward Prediction for Physical Reasoning | - |
| Dynamics-Aware DQN | 39.9 | Physical Reasoning Using Dynamics-Aware Models | - |
| DQN | 36.8 | PHYRE: A New Benchmark for Physical Reasoning | - |
0 of 4 row(s) selected.