Jailbreak Attack on MM-SafetyBench Claude-4.5 (ALL)
2.53Average ToxicityQR
Evaluation Results
| Method | Links | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| QRAttack=QR2026.04 | 2.53 | 30.09 | 8.25 | 7.98 | 18.18 | 36.81 | 49.18 | 11.69 | 11.01 | 45.1 | 19.42 | 28.46 | 39.52 | 67.89 | 47.65 | 30.09 | |
| JPSAttack=JPS2026.04 | 3.16 | 45.63 | 26.47 | 24.83 | 34.76 | 54.28 | 58.94 | 29.85 | 20.64 | 61.02 | 39.58 | 47.73 | 49.86 | 84.37 | 60.91 | 45.63 | |
| MosaicAttack=Mosaic2026.04 | 3.99 | 67.01 | 55.67 | 50.92 | 63.64 | 80.56 | 72.95 | 59.74 | 33.03 | 73.2 | 66.91 | 74.62 | 65.27 | 95.41 | 79.19 | 67.01 |