Alignment Task Evaluation on PKU-SafeRLHF w/o trigger
70.09RM Safety AccRankPoison
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| RankPoisonSetting=w/o trigger2023.11 | 70.09 | 2.02 | 14.26 | |
| Random FlipSetting=w/o trigger2023.11 | 70.05 | 1.58 | 19.6 | |
| BaselineSetting=w/o trigger2023.11 | 69.72 | 2.32 | 8.49 |