Long-context language modeling on LongBench 4-task average
12.7Average Accuracy
2d hetero
Updated 1mo ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.04 | 12.7 | 582 | |
2026.04 | 12.1 | 283 | |
2026.04 | 12 | 136 | |
2026.04 | 11.5 | 1,175 | |
2026.04 | 11.5 | 1,859 | |
2026.04 | 7.7 | 448 | |
2026.04 | 7.5 | 224 | |
2026.04 | 7.3 | 1,175 | |
2026.04 | 6.9 | 224 | |
2026.04 | 6.8 | 112 | |
2026.04 | 6.5 | 582 | |
2026.04 | 6.1 | 56 | |
2026.04 | 6 | 448 | |
2026.04 | 5.9 | 139 | |
2026.04 | 5.8 | 283 | |
2026.04 | 4.9 | 56 | |
2026.04 | 4.1 | 112 |