Mathematical Problem Solving on MATH Hard OOD (test)
96.09AccuracySkillFlow
Evaluation Results
| Method | Links | |
|---|---|---|
| SkillFlowMethod Type=Ours, Backbone Model=Qwen3.5-9B2026.05 | 96.09 | |
| FlowSteerMethod Type=Agent+RL, Backbone Model=Qwen3.5-9B2026.05 | 93.75 | |
| AgentFlowMethod Type=Agent+RL, Backbone Model=Qwen3.5-9B2026.05 | 92.19 | |
| v4-flashMethod Type=Baseline, Backbone Model=v4-flash2026.05 | 91.41 | |
| SkillRLMethod Type=Agent+RL, Backbone Model=Qwen3.5-9B2026.05 | 91.41 | |
| AFlowMethod Type=Search-based workflow, Backbone Model=Qwen3.5-9B2026.05 | 90.63 | |
| Qwen3.5Method Type=Baseline, Backbone Model=Qwen3.5-9B2026.05 | 89.06 | |
| Qwen3.5Method Type=SFT, Backbone Model=Qwen3.5-9B2026.05 | 88.28 | |
| Qwen3.5Method Type=GRPO, Backbone Model=Qwen3.5-9B2026.05 | 87.5 |