List your company
ResearchFormerly WizWand

MMD Performance and Efficiency on MMD

485Generation ScoreGreedy
-11.08117.71246.5375.29May 28, 2026
Updated 5mo ago

Evaluation results

MethodLinks
Greedythreshold (tau)=0.95, Base LLM=Qwen3-8B2026.05485656.81
Greedythreshold (tau)=0.90, Base LLM=Qwen3-8B2026.05296436.55
Greedythreshold (tau)=0.80, Base LLM=Qwen3-8B2026.05212318.46
Randthreshold (tau)=0.95, Base LLM=Qwen3-8B2026.05132199.95
OpenEvolvethreshold (tau)=0.80, Base LLM=Qwen3-8B2026.05119181.48
Randthreshold (tau)=0.90, Base LLM=Qwen3-8B2026.05119179.07
EXP3.Pthreshold (tau)=0.95, Base LLM=Qwen3-8B2026.05114172.74
Thompsonthreshold (tau)=0.95, Base LLM=Qwen3-8B2026.05106146.75
Thompsonthreshold (tau)=0.90, Base LLM=Qwen3-8B2026.05105144.75
EXP3.Pthreshold (tau)=0.90, Base LLM=Qwen3-8B2026.0595142.59
UCBthreshold (tau)=0.90, Base LLM=Qwen3-8B2026.0592117.63
UCBthreshold (tau)=0.95, Base LLM=Qwen3-8B2026.0592117.63
ShinkaEvolvethreshold (tau)=0.80, Base LLM=Qwen3-8B2026.0571112
EXP3.Pthreshold (tau)=0.80, Base LLM=Qwen3-8B2026.054052.72
Randthreshold (tau)=0.80, Base LLM=Qwen3-8B2026.052632.34
UCBthreshold (tau)=0.80, Base LLM=Qwen3-8B2026.05812.41
Thompsonthreshold (tau)=0.80, Base LLM=Qwen3-8B2026.05812.41