Mathematical Reasoning on AIME 24 (Avg@8, Pass@8)
19.17Avg@8IntroLLM
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| IntroLLMBackbone=Qwen3-4B-Base2026.02 | 19.17 | 40 | |
| TAMPOBackbone=Qwen3-4B-Base, Range=0.6-1.52026.02 | 15 | 33.33 | |
| GRPOBackbone=Qwen3-4B-Base, Temperature (τ)=1.22026.02 | 14.58 | 30 | |
| GRPOBackbone=Qwen3-4B-Base, Temperature (τ)=1.02026.02 | 14.17 | 33.33 | |
| EADBackbone=Qwen3-4B-Base, Schedule=1.2→0.12026.02 | 14.17 | 36.67 | |
| GRPOBackbone=Qwen3-4B-Base, Temperature (τ)=0.62026.02 | 11.67 | 30 | |
| IntroLLMBackbone=Qwen3-1.7B-Base2026.02 | 9.58 | 26.67 | |
| TAMPOBackbone=Qwen3-1.7B-Base, Range=0.6-1.52026.02 | 7.92 | 23.33 | |
| GRPOBackbone=Qwen3-1.7B-Base, Temperature (τ)=1.22026.02 | 7.5 | 23.33 | |
| GRPOBackbone=Qwen3-1.7B-Base, Temperature (τ)=1.02026.02 | 7.08 | 16.67 | |
| EADBackbone=Qwen3-1.7B-Base, Schedule=1.2→0.12026.02 | 7.08 | 23.33 | |
| GRPOBackbone=Qwen3-1.7B-Base, Temperature (τ)=0.62026.02 | 6.67 | 20 | |
| BaseBackbone=Qwen3-4B-Base2026.02 | 6.67 | 20 | |
| BaseBackbone=Qwen3-1.7B-Base2026.02 | 1.67 | 10 |