| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Explanation Faithfulness | Jailbreaking DAN (test) | Label Change Rate24 | 8 | |
| Explanation Faithfulness | Jailbreaking GCG (test) | Rate of Label Changes0 | 8 | |
| Jailbreaking | Jailbreaking | Best ASR97 | 3 | |
| Jailbreaking | Jailbreaking | HQ Neuron Count185 | 2 |