Mathematical Reasoning on Aggregate (AIME, AMC, MATH, Minerva, Olympiad)
48.6Average ScoreEAPO
Evaluation results
| Method | Links | |
|---|---|---|
| EAPOBackbone=Qwen2.5-Math-7B, Method Category=Policy Entropy Method2026.04 | 48.6 | |
| ForkingBackbone=Qwen2.5-Math-7B, Method Category=Policy Entropy Method2026.04 | 47.9 | |
| EntroAdvBackbone=Qwen2.5-Math-7B, Method Category=Policy Entropy Method2026.04 | 47.9 | |
| DAPOBackbone=Qwen2.5-Math-7B, Method Category=Classic RLVR Method2026.04 | 47.2 | |
| KL-CovBackbone=Qwen2.5-Math-7B, Method Category=Policy Entropy Method2026.04 | 46.4 | |
| EntroRegBackbone=Qwen2.5-Math-7B, Method Category=Policy Entropy Method2026.04 | 46.2 | |
| GRPOBackbone=Qwen2.5-Math-7B, Method Category=Classic RLVR Method2026.04 | 43.8 | |
| W-REINFBackbone=Qwen2.5-Math-7B, Method Category=Classic RLVR Method2026.04 | 41.6 | |
| PRIMEBackbone=Qwen2.5-Math-7B, Method Category=Classic RLVR Method2026.04 | 30.1 | |
| Base-7BBackbone=Qwen2.5-Math-7B, Method Category=Classic RLVR Method2026.04 | 20.8 |