List your company

Single-Hop Question Answering on PopQA 2018 Wikipedia dump (dev)

47.2AccuracyMR-Search
9.34419.1722938.828Mar 11, 2026

Evaluation results

MethodLinks
MR-SearchBackbone=Qwen2.5-7b2026.0347.2
MR-SearchBackbone=Qwen2.5-3b2026.0346
Search-R1Backbone=Qwen2.5-3b2026.0345.6
StepResearchBackbone=Qwen2.5-3b2026.0345.6
Search-R1Backbone=Qwen2.5-7b2026.0344.9
PPRMBackbone=Qwen2.5-7b2026.0343.7
StepResearchBackbone=Qwen2.5-7b2026.0343.1
ReSearchBackbone=Qwen2.5-3b2026.0343
PPRMBackbone=Qwen2.5-3b2026.0341.1
ReSearchBackbone=Qwen2.5-7b2026.0339.1
Search-o1Backbone=Qwen2.5-3b2026.0326.2
Direct InferenceBackbone=Qwen2.5-7b2026.0314
Search-o1Backbone=Qwen2.5-7b2026.0313.1
Direct InferenceBackbone=Qwen2.5-3b2026.0310.8