AdvBench & XSTest
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Harmful prompt detection | AdvBench & XSTest (held-out evaluation sets) | AUROC (Harmful/Normal)0.9642 | 6 |
Showing 1 of 1 rows
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Harmful prompt detection | AdvBench & XSTest (held-out evaluation sets) | AUROC (Harmful/Normal)0.9642 | 6 |