List your company

Open-domain Question Answering on NQ-Open OOD (test)

82.81Exact Match (EM)SkillFlow
19.442835.893952.34568.7961May 13, 2026

Evaluation results

MethodLinks
SkillFlowMethod Type=Ours, Backbone Model=Qwen3.5-9B2026.0582.8186.04
FlowSteerMethod Type=Agent+RL, Backbone Model=Qwen3.5-9B2026.0580.4784.88
AgentFlowMethod Type=Agent+RL, Backbone Model=Qwen3.5-9B2026.0578.9183.44
SkillRLMethod Type=Agent+RL, Backbone Model=Qwen3.5-9B2026.0578.9183.09
AFlowMethod Type=Search-based workflow, Backbone Model=Qwen3.5-9B2026.0575.7980.23
v4-flashMethod Type=Baseline, Backbone Model=v4-flash2026.0537.547.9
Qwen3.5Method Type=SFT, Backbone Model=Qwen3.5-9B2026.0523.4430.97
Qwen3.5Method Type=Baseline, Backbone Model=Qwen3.5-9B2026.0521.8830.03
Qwen3.5Method Type=GRPO, Backbone Model=Qwen3.5-9B2026.0521.8828.84