Red Teaming on DailyDialog against DialoGPT-large
40RSRBRT (e+r)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| BRT (e+r)Query Limit (NQ)=20,0002023.05 | 40 | 24.9 | |
| BRT (e)Query Limit (NQ)=20,0002023.05 | 37.1 | 34.5 | |
| SLQuery Limit (NQ)=20,000, Red Teamer Backbone=OPT-1.3B2023.05 | 13.1 | 49.4 | |
| SFSQuery Limit (NQ)=20,000, Red Teamer Backbone=OPT-1.3B2023.05 | 11.7 | 43.6 | |
| BRT (s+r)Query Limit (NQ)=20,0002023.05 | 5.4 | 37.7 | |
| Offensive Top-NQQuery Limit (NQ)=20,0002023.05 | 5.3 | 38.1 | |
| BRT (s)Query Limit (NQ)=20,0002023.05 | 4.9 | 38.5 | |
| RandQuery Limit (NQ)=20,0002023.05 | 1.9 | 38.8 |