Mathematical Reasoning on AIME 2024 (Accuracy and Detailed Analysis)
85AccuracyDAMI-Pref
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| DAMI-PrefModel Base Pair=Qwen3-4B-2507-Thinking/Qwen3-4B-2507-Instruct, Sampling Rounds=8x2026.01 | 85 | 15,412 | 80 | 95 | |
| DAMI-ConfModel Base Pair=Qwen3-4B-2507-Thinking/Qwen3-4B-2507-Instruct, Sampling Rounds=8x2026.01 | 85 | 13,485 | 70 | 79 | |
| DEERModel Base Pair=Qwen3-4B-2507-Thinking/Qwen3-4B-2507-Instruct, Sampling Rounds=8x2026.01 | 83.3 | 18,065 | 94 | 100 | |
| MI-07Model Base Pair=Qwen3-4B-2507-Thinking/Qwen3-4B-2507-Instruct, Sampling Rounds=8x2026.01 | 82.5 | 13,993 | 73 | 100 | |
| DAMI-PromptModel Base Pair=Qwen3-4B-2507-Thinking/Qwen3-4B-2507-Instruct, Sampling Rounds=8x2026.01 | 80 | 10,648 | 55 | 76 | |
| ThinkingModel Base Pair=Qwen3-4B-2507-Thinking/Qwen3-4B-2507-Instruct, Sampling Rounds=8x2026.01 | 78.3 | 19,241 | 100 | 100 | |
| TAModel Base Pair=Qwen3-4B-2507-Thinking/Qwen3-4B-2507-Instruct, Sampling Rounds=8x2026.01 | 78.3 | 13,515 | 70 | 98 | |
| RoutingModel Base Pair=Qwen3-4B-2507-Thinking/Qwen3-4B-2507-Instruct, Sampling Rounds=8x2026.01 | 76.7 | 19,861 | 103 | 100 | |
| MI-05Model Base Pair=Qwen3-4B-2507-Thinking/Qwen3-4B-2507-Instruct, Sampling Rounds=8x2026.01 | 73.3 | 9,321 | 48 | 50 | |
| DAMI-PredModel Base Pair=Qwen3-4B-2507-Thinking/Qwen3-4B-2507-Instruct, Sampling Rounds=8x2026.01 | 73.3 | 7,703 | 40 | 0 | |
| TIESModel Base Pair=Qwen3-4B-2507-Thinking/Qwen3-4B-2507-Instruct, Sampling Rounds=8x2026.01 | 68.3 | 8,141 | 42 | 0 | |
| MI-03Model Base Pair=Qwen3-4B-2507-Thinking/Qwen3-4B-2507-Instruct, Sampling Rounds=8x2026.01 | 67.5 | 9,708 | 50 | 0 | |
| InstructModel Base Pair=Qwen3-4B-2507-Thinking/Qwen3-4B-2507-Instruct, Sampling Rounds=8x2026.01 | 61.7 | 8,938 | 46.5 | 0 | |
| DEERModel Base Pair=DeepSeek-R1-Distill-Qwen-7B/Qwen2.5-Math-7B, Sampling Rounds=8x2026.01 | 45 | 13,945 | 78 | 100 | |
| DAMI-PrefModel Base Pair=DeepSeek-R1-Distill-Qwen-7B/Qwen2.5-Math-7B, Sampling Rounds=8x2026.01 | 45 | 14,561 | 81 | 92 | |
| DAMI-ConfModel Base Pair=DeepSeek-R1-Distill-Qwen-7B/Qwen2.5-Math-7B, Sampling Rounds=8x2026.01 | 44.2 | 13,835 | 77 | 85 | |
| ThinkingModel Base Pair=DeepSeek-R1-Distill-Qwen-7B/Qwen2.5-Math-7B, Sampling Rounds=8x2026.01 | 43.3 | 17,887 | 100 | 100 | |
| TAModel Base Pair=DeepSeek-R1-Distill-Qwen-7B/Qwen2.5-Math-7B, Sampling Rounds=8x2026.01 | 39.2 | 9,396 | 53 | 92 | |
| MI-07Model Base Pair=DeepSeek-R1-Distill-Qwen-7B/Qwen2.5-Math-7B, Sampling Rounds=8x2026.01 | 36.7 | 18,693 | 105 | 50 | |
| DAMI-PredModel Base Pair=DeepSeek-R1-Distill-Qwen-7B/Qwen2.5-Math-7B, Sampling Rounds=8x2026.01 | 26.7 | 13,209 | 74 | 31 | |
| RoutingModel Base Pair=DeepSeek-R1-Distill-Qwen-7B/Qwen2.5-Math-7B, Sampling Rounds=8x2026.01 | 25 | 1,875 | 10 | 20 | |
| MI-03Model Base Pair=DeepSeek-R1-Distill-Qwen-7B/Qwen2.5-Math-7B, Sampling Rounds=8x2026.01 | 23.3 | 3,459 | 19 | 0 | |
| InstructModel Base Pair=DeepSeek-R1-Distill-Qwen-7B/Qwen2.5-Math-7B, Sampling Rounds=8x2026.01 | 20 | 1,470 | — | 0 | |
| MI-05Model Base Pair=DeepSeek-R1-Distill-Qwen-7B/Qwen2.5-Math-7B, Sampling Rounds=8x2026.01 | 20 | 15,330 | 86 | 0 | |
| DAMI-PromptModel Base Pair=DeepSeek-R1-Distill-Qwen-7B/Qwen2.5-Math-7B, Sampling Rounds=8x2026.01 | 20 | 3,962 | 22 | 0 | |
| TIESModel Base Pair=DeepSeek-R1-Distill-Qwen-7B/Qwen2.5-Math-7B, Sampling Rounds=8x2026.01 | 16.7 | 1,739 | 10 | 0 |