Mathematical Reasoning on MATH (Avg@4, Pass@4)
91.1Avg@4 ScoreEAPO
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| EAPOMethod Category=Policy Entropy Method, Backbone=DeepScaleR-1.5B-8k2026.04 | 91.1 | 95.2 | 57 | |
| EntroAdvMethod Category=Policy Entropy Method, Backbone=DeepScaleR-1.5B-8k2026.04 | 90.2 | 94.5 | 56.3 | |
| KL-CovMethod Category=Policy Entropy Method, Backbone=DeepScaleR-1.5B-8k2026.04 | 90 | 94.2 | 56.5 | |
| DAPOMethod Category=Classic RLVR Method, Backbone=DeepScaleR-1.5B-8k2026.04 | 89.9 | 94 | 55.1 | |
| GRPOMethod Category=Classic RLVR Method, Backbone=DeepScaleR-1.5B-8k2026.04 | 88.9 | 93.4 | 53.9 | |
| ForkingMethod Category=Policy Entropy Method, Backbone=DeepScaleR-1.5B-8k2026.04 | 87.6 | 93.1 | 53.7 | |
| W-REINFMethod Category=Classic RLVR Method, Backbone=DeepScaleR-1.5B-8k2026.04 | 87.5 | 92.6 | 51.8 | |
| EntroRegMethod Category=Policy Entropy Method, Backbone=DeepScaleR-1.5B-8k2026.04 | 86.8 | 92.5 | 51.9 | |
| Base-1.5BMethod Category=Classic RLVR Method, Backbone=DeepScaleR-1.5B-8k2026.04 | 85.7 | 91.9 | 50.7 |