Security
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Tool-use security and utility evaluation | Security | Utility Rate98.4 | 16 | |
| Task Routing | Security | Cost ($)0.01 | 15 | |
| Misaligned Task Learning | Security In-domain | Misalignment2.1 | 6 | |
| Emergent Misalignment Measurement | Security General evaluation | Misalignment Score1.21 | 6 | |
| Explainable AI Performance Evaluation | Security | Composite Score (Entropy-Weighted, Domain-Modulated)2.98 | 5 | |
| Forecasting | Security | MSE107.554 | 3 | |
| Task-Efficient Routing | Security Curated Task Benchmark 1.0 (test) | Avg. Cost0.0021 | 3 |
Showing 7 of 7 rows