Llama3.1-8B + ARPO
| Method | Links | |||
|---|---|---|---|---|
2025.12 | 34.8 | - | - | |
2026.06 | 33.1 | 22 | 2.46 | |
2026.06 | 32.8 | 16 | 2.58 | |
2026.06 | 32.1 | 17.5 | 3.1 | |
2025.12 | 31.1 | - | - | |
2025.12 | 31 | - | - | |
2026.05 | 31 | - | - | |
2026.06 | 30.7 | 15.4 | 2.26 | |
2025.12 | 30.6 | - | - | |
2026.06 | 30.6 | 16.7 | 3.07 | |
2026.06 | 30.2 | 15.9 | 2.37 | |
2025.12 | 30 | - | - | |
2025.12 | 29.9 | - | - | |
2025.12 | 29.2 | - | - | |
2025.12 | 28.7 | - | - | |
2025.12 | 28.6 | - | - | |
2026.05 | 28.3 | - | - | |
2025.12 | 27.9 | - | - | |
2026.05 | 27.5 | - | - | |
2025.12 | 25.2 | - | - | |
2026.06 | 25.1 | 13.5 | 4.71 | |
2025.12 | 24.7 | - | - | |
2026.06 | 23.4 | 17.5 | 2.26 | |
2026.06 | 22.5 | 12 | 2.33 | |
2026.05 | 21.2 | - | - | |
2026.05 | 20.9 | - | - | |
2026.06 | 19.4 | 11.5 | 2.62 | |
2026.06 | 17.8 | 9.7 | 2.67 | |
2026.06 | 17.7 | 9.8 | 2.48 | |
2026.06 | 16.8 | 11 | 4.07 | |
2026.06 | 16 | 7.8 | 2.51 | |
2026.06 | 15.7 | 8.5 | 2.52 | |
2025.12 | 15.5 | - | - | |
2025.12 | 10.4 | - | - | |
2026.06 | 9.8 | 7.4 | - | |
2025.12 | 9.5 | - | - | |
2026.06 | 7.7 | 2.5 | 2.56 | |
2025.12 | 6.6 | - | - | |
2026.06 | 6.5 | 2.5 | - | |
2025.12 | 6.1 | - | - | |
2026.06 | 6.1 | 3.2 | 1.59 | |
2025.12 | 3.6 | - | - | |
2026.06 | 3.6 | 0.5 | - |