Mathematical Reasoning on AIME 2024 (pass@1, pass@2, pass@4, pass@8)
6.25Pass@1Qwen2.5-7B (Base) + RL w/ SIE-100%
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Qwen2.5-7B (Base) + RL w/ SIE-100%Base Model=Qwen2.5-7B (Base), Learning/Inference Strategy=RL w/ SIE-100%2025.09 | 6.25 | 10.31 | 15.02 | 19.41 | |
| Llama3.1-8B-Instruct + RL w/ SIE-100%Base Model=Llama3.1-8B-Instruct, Learning/Inference Strategy=RL w/ SIE-100%2025.09 | 4.58 | 7.81 | 12.12 | 17.3 | |
| Llama3.1-8B-Instruct + CoTBase Model=Llama3.1-8B-Instruct, Learning/Inference Strategy=CoT2025.09 | 3.12 | 4.97 | 7.31 | 10.89 | |
| Qwen2.5-7B (Base) + CoTBase Model=Qwen2.5-7B (Base), Learning/Inference Strategy=CoT2025.09 | 2.29 | 3.97 | 6.27 | 9.22 |