Mathematical Reasoning on MATH 500 (Average@4)
66.3Average@4CoVRL
Evaluation Results
| Method | Links | |
|---|---|---|
| CoVRLBackbone=Qwen2.5-7B-Base2025.12 | 66.3 | |
| RAVRBackbone=Qwen2.5-7B-Base, Algorithm=GRPO2025.12 | 61.2 | |
| RLPRBackbone=Qwen2.5-7B-Base, Algorithm=GRPO2025.12 | 61.2 | |
| VeriFreeBackbone=Qwen2.5-7B-Base, Algorithm=GRPO2025.12 | 59.5 | |
| LaTROBackbone=Qwen2.5-7B-Base, Algorithm=GRPO2025.12 | 59.3 | |
| JLBBackbone=Qwen2.5-7B-Base, Algorithm=GRPO2025.12 | 57.6 | |
| Base ModelBackbone=Qwen2.5-7B-Base2025.12 | 44.7 |