Physical Commonsense Reasoning on PIQA (Accuracy and Supernodes Pruned)
84Accuracy
Unpruned
Updated 1mo ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.04 | 84 | 0 | |
2026.04 | 64 | 0 | |
2026.04 | 64 | 0 | |
2026.04 | 63 | 0 | |
2026.04 | 61 | 19.6 | |
2026.04 | 56 | 24.3 | |
2026.04 | 56 | 34.6 | |
2026.04 | 51 | 41.7 |