List your company

Monolingual Question Answering on HOTPOTQA (fEM, c3Recall)

44.14fEMSearch-R1
18.618425.244231.8738.4958Apr 28, 2026

Evaluation results

MethodLinks
Search-R1Backbone=Qwen2.5-7B-Instruct, Paradigm=Post-training2026.0444.1467.17
CroSearch-R1Backbone=Qwen2.5-7B-Instruct, Paradigm=Post-training2026.0444.1467.7
SFTBackbone=Qwen2.5-7B-Instruct, Paradigm=Post-training2026.044354.61
IRCoTBackbone=Qwen2.5-7B-Instruct, Paradigm=Training-free2026.0440.8447.67
SFTBackbone=Qwen2.5-3B-Instruct, Paradigm=Post-training2026.0439.250.79
CroSearch-R1Backbone=Qwen2.5-3B-Instruct, Paradigm=Post-training2026.0438.6766.88
RAGBackbone=Qwen2.5-7B-Instruct, Paradigm=Training-free2026.0437.650.34
Search-o1Backbone=Qwen2.5-7B-Instruct, Paradigm=Training-free2026.0436.3247.62
IRCoTBackbone=Qwen2.5-3B-Instruct, Paradigm=Training-free2026.0435.4242.7
Search-R1Backbone=Qwen2.5-3B-Instruct, Paradigm=Post-training2026.0435.3562.65
RAGBackbone=Qwen2.5-3B-Instruct, Paradigm=Training-free2026.0431.543.84
Search-o1Backbone=Qwen2.5-3B-Instruct, Paradigm=Training-free2026.0430.4441.68
Direct InferenceBackbone=Qwen2.5-7B-Instruct, Paradigm=Training-free2026.0426.543.83
Direct InferenceBackbone=Qwen2.5-3B-Instruct, Paradigm=Training-free2026.0419.635.97