Long-Horizon TAMP on Calvin Causal
100Success Rate (3/5)Flower*
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Flower*Param (M)=947, Fine-tuned=true2026.06 | 100 | 87.5 | 87.5 | |
| MDTParam (M)=4402026.06 | 100 | 100 | 35.9 | |
| Flower*Param (M)=947, Fine-tuned=true2026.06 | 100 | 54.7 | 51.6 | |
| ReStructParam (M)=5712026.06 | 100 | 81.3 | 76.6 | |
| ReStructParam (M)=5712026.06 | 98.4 | 93.8 | 92.2 | |
| FlowerParam (M)=9472026.06 | 96.9 | 57.8 | 57.8 | |
| FlowerParam (M)=9472026.06 | 92.2 | 42.2 | 39.1 | |
| MDTParam (M)=4402026.06 | 31.3 | 31.3 | 3.1 | |
| LCDParam (M)=682026.06 | 18.8 | 12.5 | 12.5 | |
| LCDParam (M)=682026.06 | 15.6 | 7.8 | 7.8 | |
| HULCParam (M)=472026.06 | 12.5 | 10.9 | 1.6 | |
| HULCParam (M)=472026.06 | 7.8 | 3.1 | 0 |