Commonsense Reasoning on PIQA (non-IID distribution, alpha=0.1)
74.1Accuracy
FedAlign-MoE
Updated 25d ago
Evaluation Results
| Method | Links | |
|---|---|---|
2026.03 | 74.1 | |
2026.03 | 73.25 | |
2026.03 | 72.14 | |
2026.03 | 71.36 | |
2026.03 | 71.26 | |
2026.03 | 70.08 | |
2026.03 | 69.18 | |
2026.03 | 68.78 | |
2026.03 | 68.24 | |
2026.03 | 67.92 |