Jailbreaking on WildJailbreak (WJB)
89.5ASR@1 (Qwen2.5-7B-IT)TRACE (single)
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| TRACE (single)Workflow Category=Multi-turn RL, Judge Model=GPT-4o2026.05 | 89.5 | 73 | 79 | |
| TRACE (mix)Workflow Category=Multi-turn RL, Judge Model=GPT-4o2026.05 | 85 | 68.17 | 78 | |
| SirenWorkflow Category=Multi-turn RL, Judge Model=GPT-4o2026.05 | 82 | 56 | 61 | |
| TROJailWorkflow Category=Multi-turn RL, Judge Model=GPT-4o2026.05 | 79.44 | 41.33 | 68 | |
| AutoDan-TurboWorkflow Category=Single-turn, Judge Model=GPT-4o2026.05 | 69 | 32 | 3 | |
| Jailbreak-R1Workflow Category=Single-turn, Judge Model=GPT-4o2026.05 | 62 | 41.5 | 10.5 | |
| CrescendoWorkflow Category=Multi-turn Workflow, Judge Model=GPT-4o2026.05 | 58 | 10.5 | 3.5 | |
| ActorAttackWorkflow Category=Multi-turn Workflow, Judge Model=GPT-4o2026.05 | 54.5 | 36.5 | 47.5 | |
| X-TeamingWorkflow Category=Multi-turn Workflow, Judge Model=GPT-4o2026.05 | 43.5 | 30.5 | 27 | |
| MUSE-AWorkflow Category=Multi-turn Workflow, Judge Model=GPT-4o2026.05 | 42.5 | 19.5 | 5.5 | |
| PAIRWorkflow Category=Single-turn, Judge Model=GPT-4o2026.05 | 37.5 | 31 | 4 |