Adversarial Attack on Jailbreak out
75ASR (OM)Mistral-7B-Instruct-v0.3
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Mistral-7B-Instruct-v0.3Attack=Prefilling2026.06 | 75 | 54 | |
| Mistral-7B-Instruct-v0.3 + Trajectory AlignmentAttack=Prefilling2026.06 | 63 | 3 | |
| Mistral-7B-Instruct-v0.3Attack=I-GCG2026.06 | 17 | 20 | |
| Mistral-7B-Instruct-v0.3Attack=PAIR2026.06 | 8 | 10 | |
| Mistral-7B-Instruct-v0.3 + Trajectory AlignmentAttack=I-GCG2026.06 | 2 | 0 | |
| Mistral-7B-Instruct-v0.3 + Trajectory AlignmentAttack=PAIR2026.06 | 0 | 0 |