Preference Alignment Evaluation on PRISM (test)
0.331BT Score (Mean)
Hard Panel
Updated 4d ago
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
2026.02 | 0.331 | - | 0.279 | - | |
2026.02 | 0.267 | - | 0.215 | - | |
2026.02 | 0.146 | - | 0.132 | - | |
2026.02 | 0.135 | - | 0.106 | - | |
2026.02 | -0.106 | - | -0.017 | - | |
2026.02 | -0.108 | - | -0.026 | - | |
2026.02 | -0.127 | - | -0.047 | - | |
2026.02 | -0.133 | - | -0.057 | - | |
2026.02 | -0.18 | - | -0.283 | - | |
2026.02 | -0.225 | - | -0.302 | - |