Legal Reasoning on Law (LLM-as-judge score)
34.4LLM-as-judge ScoreProbMoE
Evaluation Results
| Method | Links | |
|---|---|---|
| ProbMoEBackbone=Qwen1.5-MoE-A2.7B, Top-k=42026.06 | 34.4 | |
| Frozen RouterBackbone=Qwen1.5-MoE-A2.7B, Top-k=42026.06 | 33.01 | |
| DefaultMoEBackbone=Qwen1.5-MoE-A2.7B, Top-k=42026.06 | 31.2 | |
| DenseMixerBackbone=Qwen1.5-MoE-A2.7B, Top-k=42026.06 | 30.75 | |
| ConventionalBackbone=Qwen1.5-MoE-A2.7B, Top-k=42026.06 | 29.5 | |
| ProbMoEBackbone=OLMoE-1B-7B, Top-k=82026.06 | 29 | |
| DenseMixerBackbone=OLMoE-1B-7B, Top-k=82026.06 | 27.9 | |
| ReMoEBackbone=Qwen1.5-MoE-A2.7B, Top-k=42026.06 | 25.5 | |
| ConventionalBackbone=OLMoE-1B-7B, Top-k=82026.06 | 25 | |
| Frozen RouterBackbone=OLMoE-1B-7B, Top-k=82026.06 | 22.5 | |
| Base ModelBackbone=Qwen1.5-MoE-A2.7B, Top-k=42026.06 | 18.2 | |
| Base ModelBackbone=OLMoE-1B-7B, Top-k=82026.06 | 5.7 | |
| SparseMixerBackbone=Qwen1.5-MoE-A2.7B, Top-k=42026.06 | 3.4 |