Mathematical Reasoning on AIME 2024 (Average@32)
7.5Average@32CoVRL
Evaluation Results
| Method | Links | |
|---|---|---|
| CoVRLBackbone=Qwen2.5-7B-Base2025.12 | 7.5 | |
| RLPRBackbone=Qwen2.5-7B-Base, Algorithm=GRPO2025.12 | 6.5 | |
| RAVRBackbone=Qwen2.5-7B-Base, Algorithm=GRPO2025.12 | 6.3 | |
| VeriFreeBackbone=Qwen2.5-7B-Base, Algorithm=GRPO2025.12 | 5 | |
| JLBBackbone=Qwen2.5-7B-Base, Algorithm=GRPO2025.12 | 4.8 | |
| LaTROBackbone=Qwen2.5-7B-Base, Algorithm=GRPO2025.12 | 4 | |
| Base ModelBackbone=Qwen2.5-7B-Base2025.12 | 2.7 |