Mathematical Reasoning on AIME 23-24
1,986.5Token CountTRACES
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| TRACESthreshold=0.9, Backbone=DS-Qwen14B, Seed=422026.04 | 1,986.5 | 0.05 | 84.83 | |
| TRACESthreshold=0.8, Backbone=DS-Qwen14B, Seed=422026.04 | 4,015.3 | 0.133 | 69.34 | |
| TRACESthreshold=0.7, Backbone=DS-Qwen14B, Seed=422026.04 | 5,949.5 | 0.2 | 54.57 | |
| TRACESthreshold=0.6, Backbone=DS-Qwen14B, Seed=422026.04 | 7,983.7 | 0.3 | 39.04 | |
| TRACESthreshold=0.5, Backbone=DS-Qwen14B, Seed=422026.04 | 9,478.9 | 0.383 | 27.62 | |
| Basel. P^{(1)}_{system}Backbone=DS-Qwen14B, Seed=422026.04 | 10,181.2 | 0.517 | 22.26 | |
| TRACESthreshold=0.4, Backbone=DS-Qwen14B, Seed=422026.04 | 10,524.3 | 0.417 | 19.64 | |
| Basel. P^{(0)}_{system}Backbone=DS-Qwen14B, Seed=422026.04 | 11,095.9 | 0.5 | 15.27 | |
| Basel. P^{(3)}_{system}Backbone=DS-Qwen14B, Seed=422026.04 | 11,635.7 | 0.467 | 11.15 | |
| Basel. P^{(0)}_{user}Backbone=DS-Qwen14B, Seed=422026.04 | 12,748.4 | 0.422 | 2.66 | |
| StandardBackbone=DS-Qwen14B, Seed=422026.04 | 13,096.3 | 0.483 | — |