Question Answering on MuSiQue OOD (test)
85.16Exact Match (EM)SkillFlow
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SkillFlowMethod Type=Ours, Backbone Model=Qwen3.5-9B2026.05 | 85.16 | 89.29 | |
| AgentFlowMethod Type=Agent+RL, Backbone Model=Qwen3.5-9B2026.05 | 79.69 | 84.9 | |
| FlowSteerMethod Type=Agent+RL, Backbone Model=Qwen3.5-9B2026.05 | 79.69 | 85.22 | |
| AFlowMethod Type=Search-based workflow, Backbone Model=Qwen3.5-9B2026.05 | 78.13 | 84.38 | |
| SkillRLMethod Type=Agent+RL, Backbone Model=Qwen3.5-9B2026.05 | 75.78 | 81.38 | |
| v4-flashMethod Type=Baseline, Backbone Model=v4-flash2026.05 | 50.78 | 59.32 | |
| Qwen3.5Method Type=GRPO, Backbone Model=Qwen3.5-9B2026.05 | 42.2 | 50.3 | |
| Qwen3.5Method Type=SFT, Backbone Model=Qwen3.5-9B2026.05 | 39.1 | 47.9 | |
| Qwen3.5Method Type=Baseline, Backbone Model=Qwen3.5-9B2026.05 | 39.06 | 47.79 |