Long-form Generation on FactScore (200-word constraint)
98.9Response Coverage (%)
Greedy
Updated 1mo ago
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
2026.06 | 98.9 | 16.2 | 22.4 | 58 | |
2026.06 | 98.4 | 11.8 | 29.1 | 71 | |
2026.06 | 97.8 | 16.9 | 22.3 | 56.9 | |
2026.06 | 97.8 | 14.7 | 21.3 | 59.2 | |
2026.06 | 84.6 | 6.7 | 23.6 | 77.9 |