Mathematical Reasoning on AIME Long context 2022-2024
29.33Accuracy0.7 × RoPE theta × 16 + 0.3 × 1M-Model
Evaluation results
| Method | Links | ||
|---|---|---|---|
| 0.7 × RoPE theta × 16 + 0.3 × 1M-ModelOperation=0.7 × RoPE theta × 16 + 0.3 × 1M-Model, recipe=LCTMerge2025.05 | 29.33 | 22,919 | |
| RoPE theta × 16Operation=RoPE theta × 162025.05 | 27.56 | 17,321 | |
| RoPE theta × 1Operation=RoPE theta × 12025.05 | 13.78 | 10,418 |