Multi-hop Question Answering on 2WikiMHQA in-distribution
79.39Exact Match (EM)
CRAFT7B
Updated 3mo ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.02 | 79.39 | 85.14 | 94.1 | |
2026.02 | 70.6 | 79.41 | 97.8 | |
2026.02 | 70.35 | 79.04 | 95.95 | |
2026.02 | 70.35 | 79.35 | 98.55 | |
2026.02 | 70.05 | 81.09 | 97.1 | |
2026.02 | 65.4 | 74.61 | 93.9 | |
2026.02 | 63.15 | 72 | 82.25 | |
2026.02 | 60.8 | 70.2 | 86.15 | |
2026.02 | 55.9 | 64.77 | 74.1 | |
2026.02 | 55.68 | 64.18 | 82 | |
2026.02 | 44.22 | 51.33 | 48.96 | |
2026.02 | 41.95 | 50.67 | 60.8 | |
2026.02 | 36.3 | 43.53 | 42.7 | |
2026.02 | 24.7 | 30.94 | 26.5 | |
2026.02 | 23.15 | 28.68 | 26.85 | |
2026.02 | 21.65 | 28.33 | 31.15 | |
2026.02 | 5.05 | 9.98 | 8 |