List your company

Text Summarization on Summary (LLM-as-judge)

44.4LLM-as-judge ScoreProbMoE
0.40811.82923.2534.671Jun 1, 2026

Evaluation results

MethodLinks
ProbMoEBackbone=Qwen1.5-MoE-A2.7B, Top-k=42026.0644.4
DenseMixerBackbone=Qwen1.5-MoE-A2.7B, Top-k=42026.0641
ProbMoEBackbone=OLMoE-1B-7B, Top-k=82026.0639.29
ConventionalBackbone=Qwen1.5-MoE-A2.7B, Top-k=42026.0639
DefaultMoEBackbone=Qwen1.5-MoE-A2.7B, Top-k=42026.0638.4
Frozen RouterBackbone=Qwen1.5-MoE-A2.7B, Top-k=42026.0638.29
DenseMixerBackbone=OLMoE-1B-7B, Top-k=82026.0637.5
Frozen RouterBackbone=OLMoE-1B-7B, Top-k=82026.0636.05
ConventionalBackbone=OLMoE-1B-7B, Top-k=82026.0633.7
Base ModelBackbone=Qwen1.5-MoE-A2.7B, Top-k=42026.0628.29
ReMoEBackbone=Qwen1.5-MoE-A2.7B, Top-k=42026.0625.8
Base ModelBackbone=OLMoE-1B-7B, Top-k=82026.067.4
SparseMixerBackbone=Qwen1.5-MoE-A2.7B, Top-k=42026.062.1