Monolingual Question Answering on POPQA
65.04fEMCroSearch-R1
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CroSearch-R1Backbone=Qwen2.5-3B-Instruct, Paradigm=Post-training2026.04 | 65.04 | 83.46 | |
| SFTBackbone=Qwen2.5-7B-Instruct, Paradigm=Post-training2026.04 | 62.6 | 68.4 | |
| CroSearch-R1Backbone=Qwen2.5-7B-Instruct, Paradigm=Post-training2026.04 | 61.33 | 78.04 | |
| Search-R1Backbone=Qwen2.5-3B-Instruct, Paradigm=Post-training2026.04 | 59.96 | 78.39 | |
| SFTBackbone=Qwen2.5-3B-Instruct, Paradigm=Post-training2026.04 | 59.5 | 65.18 | |
| Search-R1Backbone=Qwen2.5-7B-Instruct, Paradigm=Post-training2026.04 | 58.2 | 70.96 | |
| Search-o1Backbone=Qwen2.5-3B-Instruct, Paradigm=Training-free2026.04 | 54.04 | 62.5 | |
| IRCoTBackbone=Qwen2.5-3B-Instruct, Paradigm=Training-free2026.04 | 52.68 | 60.44 | |
| RAGBackbone=Qwen2.5-3B-Instruct, Paradigm=Training-free2026.04 | 50.1 | 59.47 | |
| IRCoTBackbone=Qwen2.5-7B-Instruct, Paradigm=Training-free2026.04 | 31.43 | 50.22 | |
| Search-o1Backbone=Qwen2.5-7B-Instruct, Paradigm=Training-free2026.04 | 30.48 | 47.06 | |
| Direct InferenceBackbone=Qwen2.5-7B-Instruct, Paradigm=Training-free2026.04 | 24.9 | 37.06 | |
| Direct InferenceBackbone=Qwen2.5-3B-Instruct, Paradigm=Training-free2026.04 | 18.5 | 30.5 | |
| RAGBackbone=Qwen2.5-7B-Instruct, Paradigm=Training-free2026.04 | 10 | 55 |