Jailbreak Robustness on Jailbreak Benchmarks Aggregated (test)
27.21Jailbreak Success RateBase
Evaluation results
| Method | Links | |
|---|---|---|
| BaseAggregation=Mean over all models and benchmarks2026.06 | 27.21 | |
| Star1Aggregation=Mean over all models and benchmarks2026.06 | 22.75 | |
| SafePathAggregation=Mean over all models and benchmarks2026.06 | 19.95 | |
| ST-SAggregation=Mean over all models and benchmarks2026.06 | 11.71 | |
| ST-DAggregation=Mean over all models and benchmarks2026.06 | 7.79 |