Semantic Alignment on Safety Concepts Smoking
-0.0996Delta OriginalUndefended Model
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Undefended Model2026.02 | -0.0996 | — | — | — | |
| ReVision2026.02 | — | -0.0902 | 0.0094 | 0.0057 |
| Method | Links | ||||
|---|---|---|---|---|---|
| Undefended Model2026.02 | -0.0996 | — | — | — | |
| ReVision2026.02 | — | -0.0902 | 0.0094 | 0.0057 |