Total
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Backdoor Detection | Total | TPR99.03 | 9 | |
| Safety Evaluation | Total | Collision Count0 | 8 | |
| Vulnerability Discovery | Total 10 Projects Aggregated | Confirmed Vulnerabilities379 | 7 | |
| Memory-bank storage size measurement | Total | Storage Size (GiB)2.99 | 6 | |
| multi-object image editing | Total | Preference Rate61.47 | 6 | |
| Word Similarity | Total Aggregate of WordSim353, MEN, Turk, SimLex-999 | Sum of Spearman rho2.172 | 6 | |
| Image-to-Line Drawing Translation | Total Contour Drawings & Anime | User Preference Rate0.988 | 5 | |
| Transferability Estimation | Total All Benchmarks | Pearson Correlation (PCC)0.88 | 4 | |
| Safe Navigation | Total Aggregate Environments | Success Rate96 | 4 | |
| Anomaly Segmentation | Total Aggregate | False Positives16.19 | 4 | |
| Privilege escalation detection | Total | True Positives (TP)39 | 1 | |
| Mathematical Formalization | Total long-form Combined Corpora (aggregate) | Blocks1,043 | 1 |
Showing 12 of 12 rows