Procedure Planning on CrossTask T=4 (test)
34.09SRCLAD
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| CLADUse ground truth a0 and at=false2025.03 | 34.09 | 74.3 | — | 74.85 | |
| PDPPUse ground truth a0 and at=true2025.03 | 27.86 | 72.69 | — | 74.16 | |
| SCHEMAUse ground truth a0 and at=true2025.03 | 27.66 | 71.77 | — | — | |
| KEPPUse ground truth a0 and at=true2025.03 | 26.76 | 72.52 | — | 73.22 | |
| ActionDiffusionUse ground truth a0 and at=true2025.03 | 23.61 | 70.64 | — | 72.32 | |
| Ext-GAILSupervision=V2022.05 | 16.41 | 43.05 | 60.93 | — | |
| P3IVSupervision=L2022.05 | 13.4 | 44.16 | 70.01 | — | |
| DDNSupervision=V2022.05 | 5.97 | 27.1 | 48.46 | — | |
| Retrieval-BasedSupervision=-2022.05 | 3.95 | 22.22 | 36.97 | — | |
| UPNSupervision=V2022.05 | 1.19 | 21.59 | 27.85 | — | |
| UAAASupervision=-2022.05 | 0.98 | 19.86 | 27.09 | — | |
| WLTDOSupervision=-2022.05 | 0.77 | 17.92 | 26.43 | — | |
| RandomSupervision=-2022.05 | 0.01 | 0.83 | 1.66 | — | |
| RandomUse ground truth a0 and at=true2025.03 | 0 | 1 | — | 1.83 |