Mathematical Reasoning on AMC 23 (Avg@8, Pass@8)
52.5Avg@8IntroLLM
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| IntroLLMBackbone=Qwen3-4B-Base2026.02 | 52.5 | 87.5 | |
| GRPOBackbone=Qwen3-4B-Base, Temperature (τ)=1.22026.02 | 51.88 | 82.5 | |
| EADBackbone=Qwen3-4B-Base, Schedule=1.2→0.12026.02 | 51.88 | 77.5 | |
| GRPOBackbone=Qwen3-4B-Base, Temperature (τ)=0.62026.02 | 50.94 | 82.5 | |
| TAMPOBackbone=Qwen3-4B-Base, Range=0.6-1.52026.02 | 50.62 | 82.5 | |
| GRPOBackbone=Qwen3-4B-Base, Temperature (τ)=1.02026.02 | 50 | 80 | |
| IntroLLMBackbone=Qwen3-1.7B-Base2026.02 | 39.06 | 72.5 | |
| GRPOBackbone=Qwen3-1.7B-Base, Temperature (τ)=1.02026.02 | 37.5 | 67.5 | |
| EADBackbone=Qwen3-1.7B-Base, Schedule=1.2→0.12026.02 | 37.5 | 70 | |
| GRPOBackbone=Qwen3-1.7B-Base, Temperature (τ)=1.22026.02 | 35.63 | 70 | |
| GRPOBackbone=Qwen3-1.7B-Base, Temperature (τ)=0.62026.02 | 35 | 67.5 | |
| TAMPOBackbone=Qwen3-1.7B-Base, Range=0.6-1.52026.02 | 35 | 70 | |
| BaseBackbone=Qwen3-4B-Base2026.02 | 27.19 | 72.5 | |
| BaseBackbone=Qwen3-1.7B-Base2026.02 | 18.12 | 65 |