Overall on HOWTOBENCH
0.93Pearson Correlation (ρ)ToW
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| ToWCost ($)=7.342026.04 | 0.93 | 0.83 | 0.93 | |
| Average scoring (ToW w/o plan)Cost ($)=7.022026.04 | 0.89 | 0.61 | 0.82 | |
| Elaborated RubricSelection=worst, Cost ($)=1.312026.04 | 0.89 | 0.61 | 0.82 | |
| Elaborated RubricSelection=best, Cost ($)=1.312026.04 | 0.89 | 0.67 | 0.87 | |
| Elaborated Rubric + SCn=5, Cost ($)=6.532026.04 | 0.89 | 0.61 | 0.82 | |
| Elaborated Rubric + SCn=10, Cost ($)=13.172026.04 | 0.89 | 0.61 | 0.82 | |
| Auto-PlanSelection=best, Cost ($)=0.892026.04 | 0.88 | 0.67 | 0.83 | |
| Auto-Plan + SCn=5, Cost ($)=4.452026.04 | 0.88 | 0.67 | 0.83 | |
| Auto-Plan + SCn=10, Cost ($)=8.932026.04 | 0.88 | 0.61 | 0.82 | |
| Auto-PlanSelection=worst, Cost ($)=0.892026.04 | 0.87 | 0.5 | 0.62 | |
| BLEU-12026.04 | 0.75 | 0.56 | 0.72 | |
| ROUGE-L2026.04 | 0.46 | 0.06 | 0.17 | |
| BLEU-rt2026.04 | -0.19 | -0.11 | -0.2 |