| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| CWE-Prediction | Foundation-Sec-8B-Reasoning | Accuracy70.4 | 17 | 5mo ago | |
| Software Engineering (SE) | MESA | Spearman rho0.65 | 10 | 26d ago | |
| Customer Service (CS) | MESA | Spearman Rho0.73 | 9 | 26d ago | |
| Debate | MESA | Spearman Rho0.81 | 8 | 26d ago | |
| SecurityWeek news articles Manual 2000 predictions (val) | M2: Threshold-Based Detection | Validated Relevant Count1,625 | 4 | 5mo ago |
