Malicious-skill detection on 906-skill benchmark
94.6F1 Score
Behavioral integrity (BIV)
Updated 24d ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.06 | 94.6 | 97.8 | 7.2 | |
2026.06 | 93 | - | - | |
2026.06 | 73.5 | 81 | 32 | |
2026.06 | 44 | - | - |
Behavioral integrity (BIV)
| Method | Links | |||
|---|---|---|---|---|
2026.06 | 94.6 | 97.8 | 7.2 | |
2026.06 | 93 | - | - | |
2026.06 | 73.5 | 81 | 32 | |
2026.06 | 44 | - | - |