Safety and Helpfulness Evaluation on VLSBench
84.41Safety Score
SAFER-STEER
Updated 1mo ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.03 | 84.41 | 65.74 | |
2026.03 | 82.11 | 68.27 | |
2026.03 | 76.2 | 54.05 | |
2026.03 | 76.07 | 55.04 | |
2026.03 | 72.99 | 48.27 | |
2026.03 | 72.54 | 61.92 | |
2026.03 | 69.89 | 55.8 | |
2026.03 | 67.31 | 47.39 | |
2026.03 | 65.15 | 54.26 | |
2026.03 | 62.86 | 59.31 | |
2026.03 | 61.25 | 44.07 | |
2026.03 | 57.6 | 52.83 | |
2026.03 | 57.3 | 54.95 | |
2026.03 | 55.22 | 52.66 | |
2026.03 | 52.59 | 48.53 | |
2026.03 | 50.33 | 51.9 | |
2026.03 | 46.36 | 32.38 | |
2026.03 | 38.41 | 23.21 |