Jailbreak on AdvBench RPO defense (full)
92ASRTAO-Attack
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| TAO-AttackTarget Model=Llama-2-7B-Chat2026.03 | 92 | 71 | |
| I-GCGTarget Model=Llama-2-7B-Chat2026.03 | 86 | 133 |
| Method | Links | ||
|---|---|---|---|
| TAO-AttackTarget Model=Llama-2-7B-Chat2026.03 | 92 | 71 | |
| I-GCGTarget Model=Llama-2-7B-Chat2026.03 | 86 | 133 |