Mathematical Problem Solving on GSM8K
59.4Accuracy (Exact Match)DCDM (MoE)
Evaluation Results
| Method | Links | |
|---|---|---|
| DCDM (MoE)Scale=1.5B, Active Params=1.2B (2.8B), In-context examples=4-shot, Fine-tuning=true2026.05 | 59.4 | |
| DCDMScale=1.5B, Active Params=1.5B, In-context examples=4-shot, Fine-tuning=true2026.05 | 58.72 | |
| BDLMScale=1.5B, Active Params=1.5B, In-context examples=4-shot, Fine-tuning=true2026.05 | 57.13 | |
| MDLMScale=1.5B, Active Params=1.5B, In-context examples=4-shot, Fine-tuning=true2026.05 | 53.55 | |
| DCDM (MoE)Scale=0.5B, Active Params=0.4B (0.8B), In-context examples=4-shot, Fine-tuning=true2026.05 | 50.62 | |
| DCDMScale=0.5B, Active Params=0.5B, In-context examples=4-shot, Fine-tuning=true2026.05 | 50.37 | |
| BDLMScale=0.5B, Active Params=0.5B, In-context examples=4-shot, Fine-tuning=true2026.05 | 50.21 | |
| MDLMScale=0.5B, Active Params=0.5B, In-context examples=4-shot, Fine-tuning=true2026.05 | 45.32 |