Mathematical Reasoning on Average Reasoning Benchmarks
40.91Avg@8IntroLLM
Evaluation results
| Method | Links | ||
|---|---|---|---|
| IntroLLMBackbone=Qwen3-4B-Base2026.02 | 40.91 | 73.5 | |
| EADBackbone=Qwen3-4B-Base, Schedule=1.2→0.12026.02 | 39.66 | 68.79 | |
| GRPOBackbone=Qwen3-4B-Base, Temperature (τ)=1.22026.02 | 39.6 | 68.17 | |
| TAMPOBackbone=Qwen3-4B-Base, Range=0.6-1.52026.02 | 38.83 | 69.21 | |
| GRPOBackbone=Qwen3-4B-Base, Temperature (τ)=1.02026.02 | 38.5 | 68.44 | |
| GRPOBackbone=Qwen3-4B-Base, Temperature (τ)=0.62026.02 | 37.63 | 67.97 | |
| IntroLLMBackbone=Qwen3-1.7B-Base2026.02 | 30.47 | 61.79 | |
| EADBackbone=Qwen3-1.7B-Base, Schedule=1.2→0.12026.02 | 29.41 | 59.71 | |
| GRPOBackbone=Qwen3-1.7B-Base, Temperature (τ)=1.22026.02 | 29.36 | 59.58 | |
| GRPOBackbone=Qwen3-1.7B-Base, Temperature (τ)=1.02026.02 | 28.88 | 56.06 | |
| TAMPOBackbone=Qwen3-1.7B-Base, Range=0.6-1.52026.02 | 28.28 | 59.38 | |
| GRPOBackbone=Qwen3-1.7B-Base, Temperature (τ)=0.62026.02 | 27.72 | 57.7 | |
| BaseBackbone=Qwen3-4B-Base2026.02 | 22.38 | 59.63 | |
| BaseBackbone=Qwen3-1.7B-Base2026.02 | 13.19 | 49.93 |