Writing flaw detection on BenchMarker Writing - Out of Domain, Human
92.6Accuracy
Gemma-3
Updated 4d ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.02 | 92.6 | 2.6 | 0.023 | |
2026.02 | 92.6 | 52.4 | 0.484 | |
2026.02 | 92.5 | 0 | 0 | |
2026.02 | 92.3 | 11.2 | 0.093 | |
2026.02 | 90.8 | 16.3 | 0.122 | |
2026.02 | 90.2 | 5.6 | 0.015 | |
2026.02 | 88.5 | 36.9 | 0.308 | |
2026.02 | 88.1 | 35.9 | 0.296 | |
2026.02 | 87.8 | 20.5 | 0.139 | |
2026.02 | 87.6 | 32.2 | 0.256 | |
2026.02 | 87.1 | 36.5 | 0.299 | |
2026.02 | 85.9 | 39.4 | 0.325 | |
2026.02 | 85.9 | 21.8 | 0.143 | |
2026.02 | 85.3 | 38.3 | 0.312 | |
2026.02 | 84.1 | 32.2 | 0.245 | |
2026.02 | 83.1 | 35.5 | 0.278 | |
2026.02 | 81.4 | 18.5 | 0.094 | |
2026.02 | 80.1 | 33.3 | 0.248 | |
2026.02 | 79.9 | 31.6 | 0.23 | |
2026.02 | 79.3 | 29.1 | 0.202 | |
2026.02 | 76.1 | 30.7 | 0.215 | |
2026.02 | 73.6 | 25 | 0.15 | |
2026.02 | 72.7 | 22.5 | 0.121 | |
2026.02 | 71.3 | 28.2 | 0.183 | |
2026.02 | 50 | 13 | 0 | |
2026.02 | 42.3 | 14 | 0.008 | |
2026.02 | 7.5 | 13.9 | 0 |