Question Answering on LongBench 48 items median 11K tokens
39.6F1 Score
Full cache
Updated 14d ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.05 | 39.6 | 100 | - | |
2026.05 | 34.5 | 87.2 | 0.445 | |
2026.05 | 34.5 | 87.2 | 0.445 | |
2026.05 | 33.6 | 85 | - | |
2026.05 | 33.4 | 84.4 | - | |
2026.05 | 31.4 | 79.4 | - | |
2026.05 | 30 | 75.8 | 0.673 | |
2026.05 | 20.8 | 52.7 | 0.004 | |
2026.05 | 18.7 | 47.3 | 0.002 | |
2026.05 | 10.8 | 27.2 | 0.001 |