Commonsense Reasoning on Winogrande (ACC)
80.71Accuracy
LoPT-GRPO
Updated 2mo ago
Evaluation Results
| Method | Links | |
|---|---|---|
2026.05 | 80.71 | |
2026.05 | 80.35 | |
2026.05 | 80.27 | |
2026.05 | 55.64 | |
2026.05 | 54.85 | |
2026.05 | 54.3 | |
2026.05 | 52.95 | |
2026.05 | 52.88 | |
2026.05 | 52.64 | |
2026.05 | 52.49 | |
2026.05 | 52.17 | |
2026.05 | 51.85 | |
2026.05 | 51.46 | |
2026.05 | 51.3 | |
2026.05 | 51.07 | |
2026.05 | 51.07 | |
2026.05 | 51.06 | |
2026.05 | 50.98 | |
2026.05 | 50.91 | |
2026.05 | 50.83 | |
2026.05 | 50.75 | |
2026.05 | 50.67 | |
2026.05 | 50.43 | |
2026.05 | 48.15 |