Mathematical Problem Solving on AIME 2024 (pass@8)
55.5pass@8VI-CuRL
Evaluation Results
| Method | Links | |
|---|---|---|
| VI-CuRLModel=Qwen2.5-Math-7B, Verifier Setting=Majority Vote (w/o. Verifier)2026.02 | 55.5 | |
| No CurriculumModel=Qwen2.5-Math-7B, Verifier Setting=Majority Vote (w/o. Verifier)2026.02 | 53.8 | |
| VI-CuRLModel=Qwen2.5-Math-7B, Verifier Setting=Entropy (w/o. Verifier)2026.02 | 48.4 | |
| Base (No RL)Model=Qwen2.5-Math-7B, Verifier Setting=None2026.02 | 36.2 | |
| No CurriculumModel=Qwen2.5-Math-7B, Verifier Setting=Entropy (w/o. Verifier)2026.02 | 28.8 | |
| No CurriculumModel=Llama3.2-3B-Instruct, Verifier Setting=Majority Vote (w/o. Verifier)2026.02 | 23.3 | |
| VI-CuRLModel=Llama3.2-3B-Instruct, Verifier Setting=Entropy (w/o. Verifier)2026.02 | 20.8 | |
| VI-CuRLModel=Llama3.2-3B-Instruct, Verifier Setting=Majority Vote (w/o. Verifier)2026.02 | 20.4 | |
| Base (No RL)Model=Llama3.2-3B-Instruct, Verifier Setting=None2026.02 | 18.8 | |
| No CurriculumModel=Llama3.2-3B-Instruct, Verifier Setting=Entropy (w/o. Verifier)2026.02 | 5 |