List your company

Multi-hop Question Answering on 2Wiki 18

43.6Exact Match (EM)Esearch-base
-0.0811.2622.633.94Oct 1, 2025

Evaluation results

MethodLinks
Esearch-baseModel Backbone=Qwen2.5-7b, Search Engine=Wiki-18, Model Variant=Base2025.1043.651.3
Esearch-baseModel Backbone=Qwen2.5-3b, Search Engine=Wiki-18, Model Variant=Base2025.1042.849.9
Esearch-instructModel Backbone=Qwen2.5-7b, Search Engine=Wiki-18, Model Variant=Instruct2025.1041.949.4
Esearch-instructModel Backbone=Qwen2.5-3b, Search Engine=Wiki-18, Model Variant=Instruct2025.1041.550
R1-instructModel Backbone=Qwen2.5-7b, Search Engine=Wiki-18, Model Variant=Instruct2025.1030.436.1
SFTModel Backbone=Qwen2.5-7b, Search Engine=Wiki-182025.1027.732.8
Direct InferenceModel Backbone=Qwen2.5-3b, Search Engine=Wiki-182025.1026.330.8
R1-baseModel Backbone=Qwen2.5-3b, Search Engine=Wiki-18, Model Variant=Base2025.1026.231.7
R1-instructModel Backbone=Qwen2.5-3b, Search Engine=Wiki-18, Model Variant=Instruct2025.1023.932.7
Direct InferenceModel Backbone=Qwen2.5-7b, Search Engine=Wiki-182025.1023.831.9
Search-o1Model Backbone=Qwen2.5-3b, Search Engine=Wiki-182025.102329.3
R1-baseModel Backbone=Qwen2.5-7b, Search Engine=Wiki-18, Model Variant=Base2025.1022.931.5
RAGModel Backbone=Qwen2.5-7b, Search Engine=Wiki-182025.1022.130.7
RAGModel Backbone=Qwen2.5-3b, Search Engine=Wiki-182025.1019.227.1
Search-o1Model Backbone=Qwen2.5-7b, Search Engine=Wiki-182025.1018.924.6
SFTModel Backbone=Qwen2.5-3b, Search Engine=Wiki-182025.1015.824.3
IRCoTModel Backbone=Qwen2.5-3b, Search Engine=Wiki-182025.1013.719.7
IRCoTModel Backbone=Qwen2.5-7b, Search Engine=Wiki-182025.1013.321.8
CoTModel Backbone=Qwen2.5-7b, Search Engine=Wiki-182025.1012.718.4
CoTModel Backbone=Qwen2.5-3b, Search Engine=Wiki-182025.101.69.4