List your company

Multi-hop QA on MuSiQue Wiki18 index (test)

36.1Pass@4Gemma3-12B
13.63619.46825.331.132Jun 15, 2026

Evaluation results

MethodLinks
Gemma3-12BSampling Method=DivInit (DI)2026.0636.1
Gemma3-12BSampling Method=Standard parallel sampling (S)2026.0631.6
Qwen3-8BSampling Method=DivInit (DI)2026.0629.7
Qwen3-8BSampling Method=Standard parallel sampling (S)2026.0623.9
Qwen3-4BSampling Method=DivInit (DI)2026.0619.7
Gemma3-4BSampling Method=Standard parallel sampling (S)2026.0617.2
Gemma3-4BSampling Method=DivInit (DI)2026.0616.1
Qwen3-4BSampling Method=Standard parallel sampling (S)2026.0615.9
Qwen3-1.7BSampling Method=DivInit (DI)2026.0615.6
Qwen3-1.7BSampling Method=Standard parallel sampling (S)2026.0614.5