malicious-prompts
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Jailbreak Attack | Malicious Prompts English (test) | ASR@5100 | 72 | |
| Adversarial Attack | 16 malicious prompts | ASR0 | 40 | |
| Text-to-Image Generation | Malicious Prompts | FID-Censored372.38 | 6 | |
| Malicious Prompt Detection | ahsanayub/malicious-prompts | Accuracy98.72 | 4 |
Showing 4 of 4 rows