Mathematical Problem Solving on AIME 2025 (pass@8)
31.7pass@8VI-CuRL
Evaluation Results
| Method | Links | |
|---|---|---|
| VI-CuRLModel=Qwen2.5-Math-7B, Verifier Setting=Majority Vote (w/o. Verifier)2026.02 | 31.7 | |
| VI-CuRLModel=Qwen2.5-Math-7B, Verifier Setting=Entropy (w/o. Verifier)2026.02 | 30 | |
| No CurriculumModel=Qwen2.5-Math-7B, Verifier Setting=Majority Vote (w/o. Verifier)2026.02 | 28.7 | |
| Base (No RL)Model=Qwen2.5-Math-7B, Verifier Setting=None2026.02 | 23.3 | |
| No CurriculumModel=Qwen2.5-Math-7B, Verifier Setting=Entropy (w/o. Verifier)2026.02 | 17.9 | |
| VI-CuRLModel=Llama3.2-3B-Instruct, Verifier Setting=Entropy (w/o. Verifier)2026.02 | 5.8 | |
| No CurriculumModel=Llama3.2-3B-Instruct, Verifier Setting=Majority Vote (w/o. Verifier)2026.02 | 4.6 | |
| VI-CuRLModel=Llama3.2-3B-Instruct, Verifier Setting=Majority Vote (w/o. Verifier)2026.02 | 4.6 | |
| Base (No RL)Model=Llama3.2-3B-Instruct, Verifier Setting=None2026.02 | 3.3 | |
| No CurriculumModel=Llama3.2-3B-Instruct, Verifier Setting=Entropy (w/o. Verifier)2026.02 | 0 |