Mathematical Reasoning on AMC 2023 (Pass@32)
97.7Pass@32OLMo-32B-Instruct (Teacher)
Evaluation Results
| Method | Links | |
|---|---|---|
| OLMo-32B-Instruct (Teacher)Model Category=Off-the-shelf2026.06 | 97.7 | |
| SEAD (full)Model Category=Proposed (SEAD components)2026.06 | 89.8 | |
| Token Zones + KL AnnealingModel Category=Proposed (SEAD components)2026.06 | 89.1 | |
| KL Annealing onlyModel Category=Proposed (SEAD components)2026.06 | 82.9 | |
| GRPOModel Category=Baseline2026.06 | 81 | |
| OPD (RKL k=1)Model Category=Baseline2026.06 | 80.9 | |
| Token Zones onlyModel Category=Proposed (SEAD components)2026.06 | 80.9 | |
| OLMo-7B-Instruct (Student)Model Category=Off-the-shelf2026.06 | 80.4 | |
| OPSDModel Category=Baseline2026.06 | 80.2 |