List your company

Multi-hop Question Answering on MuSiQue online Google Search API (val test)

26.5EMESearch-base
9.96414.25718.5522.843Oct 1, 2025

Evaluation results

MethodLinks
ESearch-baseModel Scale=7B2025.1026.537.2
ESearch-instructModel Scale=7B2025.1025.437.5
ESearch-baseModel Scale=3B2025.1024.736.7
StepSearch-baseModel Scale=7B2025.1023.232.5
StepSearch-instructModel Scale=7B2025.1022.232.7
Research-instructModel Scale=7B2025.102232.1
Search-R1-baseModel Scale=7B2025.1021.931
ESearch-instructModel Scale=3B2025.1021.131.1
ZeroSearch-baseModel Scale=7B2025.1021.129.7
StepSearch-baseModel Scale=3B2025.1019.629.1
Search-R1-instructModel Scale=7B2025.1018.226.8
Research-baseModel Scale=7B2025.1017.627.5
R-Search-instruct-PPOModel Scale=7B2025.1015.824
ZeroSearch-baseModel Scale=3B2025.1015.223.7
R-Search-instruct-PPOModel Scale=3B2025.1015.223.4
StepSearch-instructModel Scale=3B2025.101524
R-Search-instruct-GRPOModel Scale=3B2025.1014.222.7
ZeroSearch-instructModel Scale=7B2025.1013.622.5
R-Search-instruct-GRPOModel Scale=7B2025.1013.422.5
Search-R1-instructModel Scale=3B2025.1012.819.5
Search-R1-baseModel Scale=3B2025.101218.4
ZeroSearch-instructModel Scale=3B2025.1011.417.6
SSRL-instructModel Scale=3B2025.1011.419.5
SSRL-instructModel Scale=7B2025.1010.618.4