Reasoning on WinoGrande Clean Setting
73.9AccuracyUSMoE
Evaluation Results
| Method | Links | |
|---|---|---|
| USMoEModel=Qwen3-30B-A3B-Thinking, Evaluation Protocol=0-shot2025.03 | 73.9 | |
| USMoEModel=Qwen3-30B-A3B-Instruct, Evaluation Protocol=0-shot2025.03 | 73.6 | |
| OriginalModel=Qwen3-30B-A3B-Instruct, Selection Strategy=Token Choice (TC), Evaluation Protocol=0-shot2025.03 | 73.3 | |
| OriginalModel=Qwen3-30B-A3B-Thinking, Selection Strategy=Token Choice (TC), Evaluation Protocol=0-shot2025.03 | 72.9 | |
| Expert Choice (EC)Model=Qwen3-30B-A3B-Thinking, Evaluation Protocol=0-shot2025.03 | 59.6 | |
| Expert Choice (EC)Model=Qwen3-30B-A3B-Instruct, Evaluation Protocol=0-shot2025.03 | 58.3 | |
| MoEEModel=Qwen3-30B-A3B-Thinking, Selection Strategy=Token Choice (TC), Evaluation Protocol=0-shot2025.03 | 50.5 | |
| MoEEModel=Qwen3-30B-A3B-Instruct, Selection Strategy=Token Choice (TC), Evaluation Protocol=0-shot2025.03 | 49.7 |