Multilingual Evaluation on GSM8K, IFEval, MMLU Aggregate
40.9Average ScoreRA-MoE
Evaluation Results
| Method | Links | |
|---|---|---|
| RA-MoEModel=Qwen1.52026.05 | 40.9 | |
| RISEModel=Qwen1.52026.05 | 39.9 | |
| SFTModel=Qwen1.52026.05 | 39.3 | |
| RA-MoEModel=DSV22026.05 | 39.1 | |
| RISEModel=DSV22026.05 | 38.3 | |
| SFTModel=DSV22026.05 | 38.1 | |
| RSModel=DSV22026.05 | 34.9 | |
| 0-shotModel=DSV22026.05 | 34.6 | |
| RSModel=Qwen1.52026.05 | 29.8 | |
| 0-shotModel=Qwen1.52026.05 | 29.5 | |
| RA-MoEModel=OLMoE2026.05 | 24.2 | |
| RISEModel=OLMoE2026.05 | 23.5 | |
| SFTModel=OLMoE2026.05 | 23.3 | |
| RSModel=OLMoE2026.05 | 20.7 | |
| 0-shotModel=OLMoE2026.05 | 20.4 |