Long-context Understanding on RULER 8K context
54.15CWE
Training–Inference Consistent Segmented Execution
Updated 21d ago
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| 54.15 | - | - | 38.67 | - | ||
2026.05 | 39.4 | - | - | 36.17 | - | |
| 39.25 | - | - | 35.83 | - | ||
2026.05 | 38.8 | - | - | 42.67 | - | |
2026.05 | 36.25 | - | - | 39.5 | - | |
2024.09 | 24.1 | 0.19 | 0 | 17 | 28.5 | |
2024.09 | 21.1 | 60.66 | 8.6 | 12 | 5.5 | |
2026.05 | 20.45 | - | - | 38.83 | - | |
2024.09 | 13.4 | 38.19 | 2.2 | 10.33 | 44 | |
2024.09 | 10.3 | 42.21 | 1.8 | 11.67 | 34.5 | |
2024.09 | 6.3 | 3.5 | 0.6 | 11.67 | 31 | |
2024.09 | 5.7 | 30.44 | 4.2 | 24.33 | 16 | |
2024.09 | 1.5 | 9.5 | 0.6 | 9.33 | 33.5 |