List your company

Reasoning on OBQA Corrupt Setting

34.4AccuracyUSMoE
29.230.5531.933.25Mar 29, 2025

Evaluation results

MethodLinks
USMoEModel=Qwen3-30B-A3B-Thinking, Evaluation Protocol=0-shot2025.0334.4
USMoEModel=Qwen3-30B-A3B-Instruct, Evaluation Protocol=0-shot2025.0333.6
Expert Choice (EC)Model=Qwen3-30B-A3B-Instruct, Evaluation Protocol=0-shot2025.0332.6
Expert Choice (EC)Model=Qwen3-30B-A3B-Thinking, Evaluation Protocol=0-shot2025.0332.2
MoEEModel=Qwen3-30B-A3B-Instruct, Selection Strategy=Token Choice (TC), Evaluation Protocol=0-shot2025.0330.2
MoEEModel=Qwen3-30B-A3B-Thinking, Selection Strategy=Token Choice (TC), Evaluation Protocol=0-shot2025.0330
OriginalModel=Qwen3-30B-A3B-Instruct, Selection Strategy=Token Choice (TC), Evaluation Protocol=0-shot2025.0330
OriginalModel=Qwen3-30B-A3B-Thinking, Selection Strategy=Token Choice (TC), Evaluation Protocol=0-shot2025.0329.4