List your company
ResearchFormerly WizWand

Multi-hop Question Answering on MuSiQue (pass@1)

32pass@1Qwen2.5-32B-Instruct + SSP
10.1615.8321.527.17Oct 21, 2025
Updated 5mo ago

Evaluation results

MethodLinks
Qwen2.5-32B-Instruct + SSPTraining Paradigm=Scaling to Larger Models, SSP Integration=true2025.1032
R-Search-7B + SSPTraining Paradigm=Continual Training on Search-Specialized Agents, SSP Integration=true2025.1031.4
Search-R1-7B + SSPTraining Paradigm=Continual Training on Search-Specialized Agents, SSP Integration=true2025.1030.6
R-Search-7BTraining Paradigm=Continual Training on Search-Specialized Agents, SSP Integration=false2025.1029.8
Search-R1-7BTraining Paradigm=Continual Training on Search-Specialized Agents, SSP Integration=false2025.1029.6
Qwen3-8B + SSPTraining Paradigm=Generalization Across Model Families, SSP Integration=true2025.1028
Qwen2.5-14B-Instruct + SSPTraining Paradigm=Scaling to Larger Models, SSP Integration=true2025.1028
Qwen2.5-32B-InstructTraining Paradigm=Scaling to Larger Models, SSP Integration=false2025.1027.4
Qwen3-8BTraining Paradigm=Generalization Across Model Families, SSP Integration=false2025.1026.6
Qwen2.5-14B-InstructTraining Paradigm=Scaling to Larger Models, SSP Integration=false2025.1026.6
Qwen2.5-7B-Base + SSPTraining Paradigm=From-Scratch Training on Base and Instruct Models, SSP Integration=true2025.1024
Qwen2.5-7B-Instruct + SSPTraining Paradigm=From-Scratch Training on Base and Instruct Models, SSP Integration=true2025.1021.2
ZeroSearch-7B + SSPTraining Paradigm=Continual Training on Search-Specialized Agents, SSP Integration=true2025.1019.6
ZeroSearch-7BTraining Paradigm=Continual Training on Search-Specialized Agents, SSP Integration=false2025.1017.6
Qwen2.5-7B-InstructTraining Paradigm=From-Scratch Training on Base and Instruct Models, SSP Integration=false2025.1016.8
LLaMA-3.1-8B + SSPTraining Paradigm=Generalization Across Model Families, SSP Integration=true2025.1016.2
LLaMA-3.1-8BTraining Paradigm=Generalization Across Model Families, SSP Integration=false2025.1011.4
Qwen2.5-7B-BaseTraining Paradigm=From-Scratch Training on Base and Instruct Models, SSP Integration=false2025.1011