List your company

Question Answering on MuSiQue OOD (test)

85.16Exact Match (EM)SkillFlow
37.21649.66362.1174.557May 13, 2026

Evaluation results

MethodLinks
SkillFlowMethod Type=Ours, Backbone Model=Qwen3.5-9B2026.0585.1689.29
AgentFlowMethod Type=Agent+RL, Backbone Model=Qwen3.5-9B2026.0579.6984.9
FlowSteerMethod Type=Agent+RL, Backbone Model=Qwen3.5-9B2026.0579.6985.22
AFlowMethod Type=Search-based workflow, Backbone Model=Qwen3.5-9B2026.0578.1384.38
SkillRLMethod Type=Agent+RL, Backbone Model=Qwen3.5-9B2026.0575.7881.38
v4-flashMethod Type=Baseline, Backbone Model=v4-flash2026.0550.7859.32
Qwen3.5Method Type=GRPO, Backbone Model=Qwen3.5-9B2026.0542.250.3
Qwen3.5Method Type=SFT, Backbone Model=Qwen3.5-9B2026.0539.147.9
Qwen3.5Method Type=Baseline, Backbone Model=Qwen3.5-9B2026.0539.0647.79