Procedure Planning on CrossTask T=6 (test)
12.66Success RateCLAD
Evaluation Results
| Method | Links | |
|---|---|---|
| CLADUse ground truth a0 and at=false2025.03 | 12.66 | |
| SCHEMAUse ground truth a0 and at=true2025.03 | 10.1 | |
| PDPPUse ground truth a0 and at=true2025.03 | 9.98 | |
| KEPPUse ground truth a0 and at=true2025.03 | 9.59 | |
| LAPData curation setting=KEPP setting2026.03 | 8.79 | |
| LAP-TextData curation setting=KEPP setting, Input modality=text description2026.03 | 8.66 | |
| ActionDiffusionUse ground truth a0 and at=true2025.03 | 8.13 | |
| PDPPData curation setting=KEPP setting2026.03 | 8.07 | |
| ActionDiffusionData curation setting=KEPP setting2026.03 | 8.07 | |
| KEPPData curation setting=KEPP setting2026.03 | 7.89 | |
| SCHEMAData curation setting=KEPP setting2026.03 | 7.54 | |
| SkipPlanData curation setting=KEPP setting2026.03 | 5.24 | |
| RandomUse ground truth a0 and at=true2025.03 | 0 |