Honesty Alignment on HonestyBench In-Domain
85.16NQ Score
EliCal
Updated 3mo ago
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
2025.10 | 85.16 | 89.09 | 86.09 | 84.19 | 88.89 | 86.49 | |
2025.10 | 84.89 | 88.96 | 85.64 | 83.97 | 88.07 | 86.2 | |
2025.10 | 82.38 | 87.51 | 84.48 | 82.05 | 84.31 | 84.36 | |
2025.10 | 80.68 | 90.2 | 80.12 | 55.4 | 62.93 | 73.62 | |
2025.10 | 77.86 | 86.23 | 77.27 | 54.36 | 62.05 | 71.19 | |
2025.10 | 72.19 | 68.75 | 74.34 | 76.17 | 78.61 | 73.41 | |
2025.10 | 68.82 | 62.35 | 70.53 | 73.24 | 71.5 | 68.9 | |
2025.10 | 66.11 | 72.96 | 61.96 | 59.33 | 61.67 | 64.75 | |
2025.10 | 65.02 | 74.98 | 68.98 | 67.82 | 66.35 | 69.8 | |
2025.10 | 64.02 | 70.22 | 66.49 | 65.02 | 70.81 | 67.22 | |
2025.10 | 61.69 | 72.54 | 66.39 | 65.03 | 71.06 | 67.7 | |
2025.10 | 58.15 | 63.38 | 58.08 | 55.24 | 67.58 | 59.48 | |
2025.10 | 56.79 | 70.26 | 54.29 | 41.73 | 58.71 | 55.48 |