| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Inference Latency | JailbreakV | Latency (s)2.81 | 25 | |
| Text-Based Jailbreak Attack | JailbreakV-28K (test) | ASR (None-Template)75.23 | 25 | |
| Safety Evaluation | JailbreakV-28K v1 (test) | ASR (Noise-T)6.63 | 18 | |
| Safety Evaluation | JailBreakV | ASR6.55 | 18 | |
| Abnormal Behavior Detection | JailBreakV (test) | Accuracy100 | 17 | |
| Malicious Prompt Detection | JailbreakV_28K Text-based (test) | FNR0 | 16 | |
| Malicious Prompt Detection | JailbreakV_28K Image-based (test) | FNR0.19 | 16 | |
| Image-based Jailbreak | JailbreakV_28K IND | ASR0.19 | 16 | |
| Response Safety | JailBreakV-28K (avg) | JBV-R Score0.975 | 15 | |
| Jailbreak Defense | JailBreakV | Attack Success Rate (ASR)6.55 | 14 | |
| Transfer Attack | JailBreakV-28K | ASR (with SN)78.9 | 11 | |
| Jailbreak Attack | JailbreakV | ASR0 | 10 | |
| Safety Evaluation | JailbreakV-28K MLLM | ASR0 | 10 | |
| Safety Evaluation | JailbreakV-28K LLM | ASR1.46 | 10 | |
| Jailbreak Detection | JailbreakV | AUROC99.69 | 9 | |
| Jailbreak Defense | JailbreakV-28K | ASR (Noise, T)8.4 | 6 | |
| Jailbreak Attack Defense | JailbreakV-28K v1 (test) | Defense Success Rate (Noise - T)38.16 | 6 | |
| Jailbreak Attack | JailBreakV_28K | Attack Success Rate (ASR)58.97 | 3 | |
| Robustness to Jailbreak Attacks | JailbreakV-28K | Harmful Reasoning Ratio24.5 | 3 | |
| Safety Evaluation | JailBreakV (test) | HPR23 | 3 |