Logical Reasoning on RP Distribution, Npred ≤ 60, δ ≤ 6 (train)
99.5Accuracydirect w. L = 128
Evaluation results
| Method | Links | |
|---|---|---|
| direct w. L = 128Reasoning Strategy=Direct, Number of Layers=1282026.05 | 99.5 | |
| cot w. L = 64Reasoning Strategy=CoT, Number of Layers=642026.05 | 98.4 | |
| cot w. L = 128Reasoning Strategy=CoT, Number of Layers=1282026.05 | 97 | |
| direct (baseline)Reasoning Strategy=Direct2026.05 | 96.9 | |
| cot (baseline)Reasoning Strategy=CoT2026.05 | 95.4 |