Plan Generation on WikiHow (test)
56.1ROUGE-1Self-Ask
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Self-Askreveal_range=k in [1, 4]2026.01 | 56.1 | 47.4 | 26 | |
| ReActreveal_range=k in [1, 4]2026.01 | 55.8 | 47.4 | 76.9 | |
| ToSreveal_range=k in [1, 4]2026.01 | 53.3 | 45.4 | 82.6 | |
| ToTreveal_range=k in [1, 4]2026.01 | 52.9 | 45.2 | 94.7 | |
| SQ-BCPreveal_range=k in [1, 4]2026.01 | 52.7 | 45.9 | 14.9 | |
| CoTreveal_range=k in [1, 4]2026.01 | 48.5 | 44.7 | 83.2 | |
| Direct Promptreveal_range=k in [1, 4]2026.01 | 46.3 | 42.1 | 78.3 |