Multi-hop Question Answering on HotpotQA (Avg@8 accuracy)
92.52Avg@8 Accuracy
ST-EVO
Updated 4d ago
Evaluation Results
| Method | Links | |
|---|---|---|
2026.02 | 92.52 | |
2026.02 | 90.04 | |
2026.02 | 89.35 | |
2026.02 | 88.29 | |
2026.02 | 87.66 | |
2026.02 | 87.22 | |
2026.02 | 86.45 | |
2026.02 | 85.4 | |
2026.02 | 84.92 | |
2026.02 | 84.55 | |
2026.02 | 84.28 | |
2026.02 | 83.52 | |
2026.02 | 83.35 | |
2026.02 | 83.25 | |
2026.02 | 82.35 | |
2026.02 | 82.24 | |
2026.02 | 81.5 | |
2026.02 | 81.04 | |
2026.02 | 80.77 | |
2025.12 | 68 | |
2025.12 | 67 | |
2025.12 | 63 | |
2025.12 | 62 | |
2025.12 | 58 | |
2025.12 | 52 | |
2026.01 | 49 | |
2026.01 | 46.44 | |
2026.01 | 37.44 | |
2026.01 | 36.88 | |
2026.01 | 36.63 | |
2026.01 | 35.5 | |
2026.01 | 33.56 |