Long-context evaluation on RULER 128k Zero-shot transfer
31.3Accuracy (RULER 128k Zero-shot)
Still
Updated 1mo ago
Evaluation Results
| Method | Links | |
|---|---|---|
2026.06 | 31.3 | |
2026.06 | 30.4 | |
2026.06 | 25.8 | |
2026.06 | 23.4 | |
2026.06 | 22 | |
2026.06 | 20.7 | |
2026.06 | 20.5 | |
2026.06 | 20.3 | |
2026.06 | 19.8 | |
2026.06 | 19.5 | |
2026.06 | 18.9 | |
2026.06 | 18.1 |