Physical Reasoning on PHYRE Cross-template 1.0
50.86Success RateRPIN
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| RPINMechanism=LfD, Input=Initial scenes, bboxs, Supervision=Bboxs, masks, outcomes2022.10 | 50.86 | — | |
| Dec [Joint]Conditioning frames (tau)=1, Roll-out duration (tau')=02020.06 | 40.1 | 40.7 | |
| Dec [Joint]Conditioning frames (tau)=1, Roll-out duration (tau')=102020.06 | 39.2 | 40.3 | |
| DQNConditioning frames (tau)=12020.06 | 34.5 | 36.8 | |
| MEMConditioning frames (tau)=12020.06 | 15.2 | 18.5 | |
| RANDConditioning frames (tau)=12020.06 | 6.8 | 13 |