Mathematical Reasoning on AMC 2023 (Token Efficiency and Baseline Comparison)
90AccuracyBaseline
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| BaselineBackbone=QwQ-32B, Temperature=0.62025.12 | 90 | — | 6,505.08 | — | |
| DEERBackbone=QwQ-32B, Temperature=0.62025.12 | 90 | 0 | 6,237 | -4.1 | |
| BaselineBackbone=QwQ-32B, Temperature=0.02025.12 | 90 | — | 6,538.27 | — | |
| LYNXBackbone=QwQ-32B, Conf=0.97, Temperature=0.62025.12 | 87.5 | -2.78 | 4,817.25 | -25.9 | |
| DEERBackbone=QwQ-32B, Temperature=0.02025.12 | 87.5 | -2.78 | 5,545.88 | -15.2 | |
| LYNXBackbone=QwQ-32B, Conf=0.95, Temperature=0.62025.12 | 85 | -5.56 | 4,551.03 | -30 | |
| LYNXBackbone=QwQ-32B, Conf=0.97, Temperature=0.02025.12 | 85 | -5.56 | 5,363.12 | -18 | |
| LYNXBackbone=QwQ-32B, Conf=0.95, Temperature=0.02025.12 | 82.5 | -8.33 | 4,828.32 | -26.2 | |
| LYNXBackbone=QwQ-32B, Conf=0.90, Temperature=0.02025.12 | 82.5 | -8.33 | 4,439.62 | -32.1 | |
| LYNXBackbone=QwQ-32B, Conf=0.9, Temperature=0.62025.12 | 77.5 | -13.89 | 4,042.7 | -37.9 | |
| LYNXBackbone=QwQ-32B, Conf=0.80, Temperature=0.02025.12 | 77.5 | -13.89 | 3,850.82 | -41.1 | |
| BaselineBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Temperature=0.62025.12 | 75 | — | 7,118.93 | — | |
| LYNXBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Conf=0.97, Temperature=0.62025.12 | 75 | 0 | 5,725.05 | -19.6 | |
| ToNBackbone=QwQ-32B, Temperature=0.62025.12 | 75 | -16.67 | 3,554.87 | -45.4 | |
| LYNXBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Conf=0.9, Temperature=0.62025.12 | 72.5 | -3.33 | 3,423.63 | -51.9 | |
| LYNXBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Conf=0.95, Temperature=0.62025.12 | 70 | -6.67 | 4,754.6 | -33.2 | |
| ToNBackbone=QwQ-32B, Temperature=0.02025.12 | 70 | -22.22 | 3,680.72 | -43.7 | |
| LYNXBackbone=QwQ-32B, Conf=0.8, Temperature=0.62025.12 | 67.5 | -25 | 3,144.23 | -51.7 | |
| DEERBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Temperature=0.62025.12 | 65 | -13.33 | 5,273.05 | -25.9 | |
| LYNXBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Conf=0.8, Temperature=0.62025.12 | 62.5 | -16.67 | 2,543.48 | -64.3 | |
| LYNXBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Conf=0.95, Temperature=0.02025.12 | 62.5 | 19.05 | 6,865.05 | -25.6 | |
| LYNXBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Conf=0.90, Temperature=0.02025.12 | 60 | 14.29 | 5,315.6 | -42.4 | |
| LYNXBackbone=QwQ-32B, Conf=0.70, Temperature=0.02025.12 | 60 | -33.33 | 2,973.38 | -54.5 | |
| LYNXBackbone=QwQ-32B, Conf=0.7, Temperature=0.62025.12 | 57.5 | -36.11 | 2,562.03 | -60.6 | |
| DEERBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Temperature=0.02025.12 | 57.5 | 9.52 | 4,846.23 | -47.5 | |
| LYNXBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Conf=0.97, Temperature=0.02025.12 | 55 | 4.76 | 7,936.6 | -14 | |
| BaselineBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Temperature=0.02025.12 | 52.5 | — | 9,226.08 | — | |
| LYNXBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Conf=0.80, Temperature=0.02025.12 | 52.5 | 0 | 4,040.78 | -56.2 | |
| ToNBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Temperature=0.62025.12 | 50 | -33.33 | 3,875.55 | -45.6 | |
| ToNBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Temperature=0.02025.12 | 50 | -4.76 | 5,360.97 | -41.9 | |
| LYNXBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Conf=0.70, Temperature=0.02025.12 | 45 | -14.29 | 3,654.85 | -60.4 | |
| LYNXBackbone=DeepSeek-R1-Distill-Qwen-1.5B, Conf=0.7, Temperature=0.62025.12 | 42.5 | -43.33 | 2,070.98 | -70.9 |