Multi-hop QA Retrieval on 2Wiki
90.1Recall@2
GFM-RAG
Updated 1mo ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.03 | 90.1 | 96 | |
2026.06 | 82.3 | 88 | |
2025.09 | 81.2 | 98.2 | |
2025.09 | 80.5 | 95.7 | |
2026.03 | 80.2 | 89.1 | |
2025.09 | 79.1 | 92.4 | |
2026.03 | 77.9 | 91.8 | |
2026.06 | 76.6 | 90.4 | |
2026.03 | 72.4 | 88.9 | |
2025.09 | 70.2 | 85.3 | |
2025.09 | 68.4 | 87 | |
2026.06 | 67.3 | 76 | |
2026.06 | 67.1 | 76.5 | |
2026.06 | 66.7 | 74.8 | |
2025.09 | 66.2 | 74.1 | |
2026.03 | 63.2 | 76.1 | |
2026.03 | 61.9 | 74.9 | |
2025.09 | 61.6 | 77.3 | |
2026.03 | 61.6 | 65.9 | |
2026.06 | 61.6 | 69 | |
2025.09 | 61.4 | 88.3 | |
2025.09 | 60.9 | 66.5 | |
2026.06 | 60.2 | 67.9 | |
2025.09 | 59.2 | 68.2 | |
2026.06 | 55.3 | 65.3 | |
2026.03 | 54.9 | 69.5 | |
2026.03 | 52.3 | 61.2 | |
2025.09 | 51.8 | 61.9 | |
2026.06 | 46.6 | 57.5 | |
2025.09 | 46.3 | 53.8 | |
2025.09 | 45.1 | 59.1 | |
2026.03 | 44.6 | 59.8 |