Jailbreak Attack on MalwareBench
81.35Attack Success Rate (ASR)CodeSpear
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CodeSpearTarget Model=MiniMax-M2.72026.06 | 81.35 | 35 | |
| CodeSpearTarget Model=MiniMax-M2.72026.06 | 81.35 | 35 | |
| CodeSpearTarget Model=MiniMax-M2.52026.06 | 80.31 | 33.75 | |
| CodeSpearTarget Model=MiniMax-M2.52026.06 | 80.31 | 33.75 | |
| CodeSpear-TTarget Model=GPT-52026.06 | 65.73 | 22.5 | |
| CodeSpearTarget Model=GPT-OSS-120B2026.06 | 64.69 | 21.88 | |
| CodeSpearTarget Model=GPT-OSS-120B2026.06 | 64.69 | 21.88 | |
| CodeSpear-TTarget Model=GPT-5-mini2026.06 | 63.12 | 24.37 | |
| CodeSpearTarget Model=GPT-5-mini2026.06 | 52.08 | 20.62 | |
| CodeSpearTarget Model=GPT-5-mini2026.06 | 52.08 | 20.62 | |
| PAIRTarget Model=MiniMax-M2.52026.06 | 51.25 | 17.71 | |
| PAIRTarget Model=MiniMax-M2.52026.06 | 51.25 | 17.71 | |
| CodeSpearTarget Model=GPT-52026.06 | 50.1 | 19.06 | |
| CodeSpearTarget Model=GPT-52026.06 | 50.1 | 19.06 | |
| PAIRTarget Model=MiniMax-M2.72026.06 | 47.5 | 18.65 | |
| PAIRTarget Model=MiniMax-M2.72026.06 | 47.5 | 18.65 | |
| CodeJailbreakerTarget Model=GPT-OSS-120B2026.06 | 45.94 | 14.06 | |
| CodeJailbreakerTarget Model=GPT-OSS-120B2026.06 | 45.94 | 14.06 | |
| CodeJailbreakerTarget Model=GPT-52026.06 | 43.44 | 11.77 | |
| CodeJailbreakerTarget Model=GPT-52026.06 | 43.44 | 11.77 | |
| PAIRTarget Model=GPT-52026.06 | 38.02 | 7.6 | |
| PAIRTarget Model=GPT-52026.06 | 38.02 | 7.6 | |
| Vanilla-TTarget Model=GPT-52026.06 | 36.35 | 24.17 | |
| Vanilla-TTarget Model=GPT-52026.06 | 36.35 | 24.17 | |
| PAIRTarget Model=GPT-OSS-120B2026.06 | 35.73 | 11.15 | |
| PAIRTarget Model=GPT-OSS-120B2026.06 | 35.73 | 11.15 | |
| PAIRTarget Model=GPT-5-mini2026.06 | 32.19 | 8.75 | |
| PAIRTarget Model=GPT-5-mini2026.06 | 32.19 | 8.75 | |
| VanillaTarget Model=GPT-52026.06 | 31.87 | 4.06 | |
| VanillaTarget Model=GPT-52026.06 | 31.87 | 4.06 | |
| APTTarget Model=MiniMax-M2.72026.06 | 30.63 | 13.02 | |
| APTTarget Model=MiniMax-M2.72026.06 | 30.63 | 13.02 | |
| CodeJailbreakerTarget Model=MiniMax-M2.52026.06 | 29.48 | 5.83 | |
| CodeJailbreakerTarget Model=MiniMax-M2.52026.06 | 29.48 | 5.83 | |
| APTTarget Model=MiniMax-M2.52026.06 | 29.38 | 10.94 | |
| APTTarget Model=MiniMax-M2.52026.06 | 29.38 | 10.94 | |
| CodeJailbreakerTarget Model=MiniMax-M2.72026.06 | 28.44 | 5.73 | |
| CodeJailbreakerTarget Model=MiniMax-M2.72026.06 | 28.44 | 5.73 | |
| APTTarget Model=GPT-52026.06 | 27.5 | 3.54 | |
| APTTarget Model=GPT-52026.06 | 27.5 | 3.54 | |
| LRLTarget Model=GPT-52026.06 | 26.56 | 4.69 | |
| LRLTarget Model=GPT-52026.06 | 26.56 | 4.69 | |
| LRLTarget Model=MiniMax-M2.52026.06 | 25.94 | 4.48 | |
| LRLTarget Model=MiniMax-M2.52026.06 | 25.94 | 4.48 | |
| APTTarget Model=GPT-5-mini2026.06 | 25.31 | 5.63 | |
| APTTarget Model=GPT-5-mini2026.06 | 25.31 | 5.63 | |
| CodeJailbreakerTarget Model=GPT-5-mini2026.06 | 25.1 | 16.25 | |
| CodeJailbreakerTarget Model=GPT-5-mini2026.06 | 25.1 | 16.25 | |
| DANTarget Model=GPT-52026.06 | 24.69 | 2.81 | |
| DANTarget Model=GPT-52026.06 | 24.69 | 2.81 | |
| VanillaTarget Model=MiniMax-M2.52026.06 | 23.75 | 3.02 | |
| VanillaTarget Model=MiniMax-M2.52026.06 | 23.75 | 3.02 | |
| LRLTarget Model=GPT-5-mini2026.06 | 23.44 | 5.31 | |
| LRLTarget Model=MiniMax-M2.72026.06 | 23.44 | 3.54 | |
| LRLTarget Model=GPT-5-mini2026.06 | 23.44 | 5.31 | |
| LRLTarget Model=MiniMax-M2.72026.06 | 23.44 | 3.54 | |
| VanillaTarget Model=GPT-5-mini2026.06 | 22.71 | 1.87 | |
| VanillaTarget Model=GPT-5-mini2026.06 | 22.71 | 1.87 | |
| VanillaTarget Model=MiniMax-M2.72026.06 | 20.31 | 2.71 | |
| VanillaTarget Model=MiniMax-M2.72026.06 | 20.31 | 2.71 | |
| Vanilla-TTarget Model=MiniMax-M2.72026.06 | 19.12 | 13.03 | |
| Vanilla-TTarget Model=MiniMax-M2.72026.06 | 19.12 | 13.03 | |
| Vanilla-TTarget Model=GPT-5-mini2026.06 | 15.83 | 13.13 | |
| Vanilla-TTarget Model=GPT-5-mini2026.06 | 15.83 | 13.13 | |
| Vanilla-TTarget Model=MiniMax-M2.52026.06 | 15.73 | 10.21 | |
| Vanilla-TTarget Model=MiniMax-M2.52026.06 | 15.73 | 10.21 | |
| LRLTarget Model=GPT-OSS-120B2026.06 | 15.1 | 2.5 | |
| LRLTarget Model=GPT-OSS-120B2026.06 | 15.1 | 2.5 | |
| Vanilla-TTarget Model=GPT-OSS-120B2026.06 | 14.27 | 12.81 | |
| Vanilla-TTarget Model=GPT-OSS-120B2026.06 | 14.27 | 12.81 | |
| DANTarget Model=MiniMax-M2.52026.06 | 12.5 | 1.56 | |
| DANTarget Model=MiniMax-M2.52026.06 | 12.5 | 1.56 | |
| APTTarget Model=GPT-OSS-120B2026.06 | 11.87 | 2.19 | |
| APTTarget Model=GPT-OSS-120B2026.06 | 11.87 | 2.19 | |
| DANTarget Model=GPT-5-mini2026.06 | 11.56 | 0.42 | |
| DANTarget Model=GPT-5-mini2026.06 | 11.56 | 0.42 | |
| VanillaTarget Model=GPT-OSS-120B2026.06 | 10.94 | 1.25 | |
| DANTarget Model=GPT-OSS-120B2026.06 | 10.94 | 1.87 | |
| DANTarget Model=MiniMax-M2.72026.06 | 10.94 | 0.63 | |
| VanillaTarget Model=GPT-OSS-120B2026.06 | 10.94 | 1.25 | |
| DANTarget Model=GPT-OSS-120B2026.06 | 10.94 | 1.87 | |
| DANTarget Model=MiniMax-M2.72026.06 | 10.94 | 0.63 |