Sponge Attack Detection on EOGen
95.77F1 ScorePerplexity-filter
Evaluation Results
| Method | Links | |
|---|---|---|
| Perplexity-filterTarget Model=LLaMA22026.01 | 95.77 | |
| SHIELDTarget Model=LLaMA22026.01 | 95.32 | |
| Harm-filterTarget Model=LLaMA22026.01 | 81.34 |
| Method | Links | |
|---|---|---|
| Perplexity-filterTarget Model=LLaMA22026.01 | 95.77 | |
| SHIELDTarget Model=LLaMA22026.01 | 95.32 | |
| Harm-filterTarget Model=LLaMA22026.01 | 81.34 |