Mathematical Reasoning on Minerva (Average@4)
25.5Average@4CoVRL
Evaluation Results
| Method | Links | |
|---|---|---|
| CoVRLBackbone=Qwen2.5-7B-Base2025.12 | 25.5 | |
| RLPRBackbone=Qwen2.5-7B-Base, Algorithm=GRPO2025.12 | 24.7 | |
| LaTROBackbone=Qwen2.5-7B-Base, Algorithm=GRPO2025.12 | 24.4 | |
| VeriFreeBackbone=Qwen2.5-7B-Base, Algorithm=GRPO2025.12 | 24 | |
| JLBBackbone=Qwen2.5-7B-Base, Algorithm=GRPO2025.12 | 23.6 | |
| RAVRBackbone=Qwen2.5-7B-Base, Algorithm=GRPO2025.12 | 23.3 | |
| Base ModelBackbone=Qwen2.5-7B-Base2025.12 | 18.6 |