LLM Alignment on Ultrafeedback
86Win Rate
Greedy SFT
Updated 1mo ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2025.09 | 86 | 0.58 | |
2025.09 | 86 | 0.68 | |
2025.09 | 85 | 0.85 | |
2025.09 | 85 | 0.87 | |
2025.09 | 85 | 1.06 | |
2025.09 | 84 | 0.98 | |
2025.09 | 83 | 1.04 | |
2025.09 | 81 | 1.27 | |
2025.09 | 80 | -0.88 | |
2025.09 | 80 | 0.22 | |
2025.09 | 79 | 0.29 | |
2025.09 | 78 | -0.69 | |
2025.09 | 78 | -0.62 | |
2025.09 | 77 | 0.47 | |
2025.09 | 73 | 0.58 | |
2025.09 | 67 | -0.19 | |
2026.06 | 27.15 | - | |
2026.06 | 26.04 | - | |
2026.06 | 26.02 | - | |
2026.06 | 25.97 | - | |
2026.06 | 25.84 | - | |
2026.06 | 25.82 | - | |
2026.06 | 25.37 | - | |
2026.06 | 24.38 | - | |
2025.09 | - | 2.21 | |
2025.09 | - | 2.22 | |
2025.09 | - | 0.34 | |
2025.09 | - | 1.19 |