Natural Language Understanding on AdvGLUE (Adversarial Evaluation)
68.29Performance Score
Ours (+GDPO)
Updated 13d ago
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
2026.05 | 68.29 | - | - | - | |
2026.05 | 67.75 | - | - | - | |
2026.05 | 66.27 | - | - | - | |
2026.05 | 64.9 | - | - | - | |
2026.05 | 59.62 | - | - | - | |
2026.05 | 58.4 | - | - | - | |
2026.05 | 58.33 | - | - | - | |
2026.05 | 57.53 | - | - | - | |
2026.04 | - | 87.3 | 63.7 | -23.6 | |
2026.04 | - | 85.1 | 74.2 | -10.9 | |
2026.04 | - | 84.7 | 73.8 | -10.9 | |
2026.04 | - | 86.9 | 77.3 | -9.6 |