Classification on HotpotQA RAG v3
90.75Macro F1
SURE-RAG calibrated
Updated 2mo ago
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
2026.05 | 90.75 | 86.62 | 95.69 | 89.94 | |
2026.05 | 89.51 | 84.65 | 95.08 | 88.78 | |
2026.05 | 88.88 | 84.96 | 94.37 | 87.31 | |
2026.05 | 83.83 | 77.88 | 93.93 | 79.67 | |
2026.05 | 65.16 | 54.44 | 73.07 | 67.98 | |
2026.05 | 64.44 | 54.17 | 71.36 | 67.78 | |
2026.05 | 61.01 | 63.94 | 67.75 | 51.33 | |
2026.05 | 59.48 | 32.03 | 79.27 | 67.15 | |
2026.05 | 35.01 | 15.58 | 30.95 | 58.48 | |
2026.05 | 18.48 | 0 | 0 | 55.45 |