List your company

Question Answering on Bamboogle (Exact Match (EM) Accuracy (%))

48EM Accuracy (%)ICRL
-1.9211.042436.96Oct 17, 2025Nov 10, 2025Dec 4, 2025Dec 28, 2025Jan 21, 2026Feb 14, 2026Mar 10, 2026

Evaluation results

MethodLinks
ICRLModel=Qwen2.5-7B2026.0348
SAPO-3B-InstructRetrieval Strategy=Agent, Backbone Model=Qwen2.5-3B2026.0343.2
SE-SearchRetrieval Strategy=Agent, Backbone Model=Qwen2.5-3B2026.0342.4
SAPO-3B-BaseRetrieval Strategy=Agent, Backbone Model=Qwen2.5-3B2026.0342.4
ParallelSearchModel=Qwen2.5-7B2026.0341.1
EvolveRBackbone Model=Qwen2.5-7B2025.1039.2
Search-R1Model=Qwen2.5-7B2026.0336.8
CriticSearchRetrieval Strategy=Agent, Backbone Model=Qwen2.5-3B2026.0336.8
Search-R1-instructBackbone Model=Qwen2.5-7B2025.1036.8
InForageRetrieval Strategy=Agent, Backbone Model=Qwen2.5-3B2026.0336
Reject SamplingModel=Qwen2.5-7B2026.0335.5
Rejection SamplingBackbone Model=Qwen2.5-7B2025.1035.5
O2-SearcherRetrieval Strategy=Agent, Backbone Model=Qwen2.5-3B2026.0334.4
AutoRefine-BaseRetrieval Strategy=Agent, Backbone Model=Qwen2.5-3B2026.0334.4
ICRLModel=Qwen2.5-3B2026.0333.6
AutoRefine-InstructRetrieval Strategy=Agent, Backbone Model=Qwen2.5-3B2026.0333.6
StepSearch-BaseRetrieval Strategy=Agent, Backbone Model=Qwen2.5-3B2026.0332.8
EvolveRBackbone Model=Qwen2.5-3B2025.1032.8
Search-o1Model=Qwen2.5-3B2026.0332
Search-o1Retrieval Strategy=Agent, Backbone Model=Qwen2.5-3B2026.0332
Search-o1Backbone Model=Qwen2.5-3B2025.1032
Search-o1Model=Qwen2.5-7B2026.0329.6
R1-baseModel=Qwen2.5-7B2026.0329.6
Search-o1Backbone Model=Qwen2.5-7B2025.1029.6
R1-baseBackbone Model=Qwen2.5-7B2025.1029.6
R1-instructModel=Qwen2.5-7B2026.0329.3
R1-instructBackbone Model=Qwen2.5-7B2025.1029.3
ZeroSearchModel=Qwen2.5-7B2026.0327.8
ReSearch-InstructRetrieval Strategy=Agent, Backbone Model=Qwen2.5-3B2026.0326.6
Search-R1Model=Qwen2.5-3B2026.0326.4
Search-R1-instructBackbone Model=Qwen2.5-3B2025.1026.4
IRCoTModel=Qwen2.5-3B2026.0324
IRCoTRetrieval Strategy=IRCoT, Backbone Model=Qwen2.5-3B2026.0324
IRCoTBackbone Model=Qwen2.5-3B2025.1024
CoTModel=Qwen2.5-7B2026.0323.2
Search-R1-InstructRetrieval Strategy=Agent, Backbone Model=Qwen2.5-3B2026.0323.2
CoTBackbone Model=Qwen2.5-7B2025.1023.2
IRCoTModel=Qwen2.5-7B2026.0322.4
R1-BaseRetrieval Strategy=None, Backbone Model=Qwen2.5-3B2026.0322.4
R1-baseBackbone Model=Qwen2.5-3B2025.1022.4
IRCoTBackbone Model=Qwen2.5-7B2025.1022.4
Reject SamplingModel=Qwen2.5-3B2026.0321
Rejection SamplingBackbone Model=Qwen2.5-3B2025.1021
RAGModel=Qwen2.5-7B2026.0320.8
RAGBackbone Model=Qwen2.5-7B2025.1020.8
R1-instructModel=Qwen2.5-3B2026.0319.2
R1-InstructRetrieval Strategy=None, Backbone Model=Qwen2.5-3B2026.0319.2
R1-instructBackbone Model=Qwen2.5-3B2025.1019.2
ZeroSearch-BaseRetrieval Strategy=Agent, Backbone Model=Qwen2.5-3B2026.0313.9
Search-R1-BaseRetrieval Strategy=Agent, Backbone Model=Qwen2.5-3B2026.0312.8
ReSearch-BaseRetrieval Strategy=Agent, Backbone Model=Qwen2.5-3B2026.0312.8
DirectModel=Qwen2.5-7B2026.0312
Direct InferenceBackbone Model=Qwen2.5-7B2025.1012
SFTModel=Qwen2.5-3B2026.0311.2
SFTModel=Qwen2.5-7B2026.0311.2
SFTRetrieval Strategy=None, Backbone Model=Qwen2.5-3B2026.0311.2
SFTBackbone Model=Qwen2.5-3B2025.1011.2
SFTBackbone Model=Qwen2.5-7B2025.1011.2
ZeroSearchModel=Qwen2.5-3B2026.0311.1
Search-R1-baseBackbone Model=Qwen2.5-3B2025.108.8
RAGModel=Qwen2.5-3B2026.038
Naive RAGRetrieval Strategy=Naive RAG, Backbone Model=Qwen2.5-3B2026.038
RAGBackbone Model=Qwen2.5-3B2025.108
DirectModel=Qwen2.5-3B2026.032.4
Direct GenerationRetrieval Strategy=None, Backbone Model=Qwen2.5-3B2026.032.4
Direct InferenceBackbone Model=Qwen2.5-3B2025.102.4
CoTModel=Qwen2.5-3B2026.030
CoTBackbone Model=Qwen2.5-3B2025.100