List your company
ResearchFormerly WizWand

Multi-step Reasoning on StrategyQA (test)

64.63AccuracyQwen3-4B + SFT + WeMask(TF)
61.634862.412463.1963.9676May 8, 2026
Updated 5mo ago

Evaluation results

MethodLinks
Qwen3-4B + SFT + WeMask(TF)Mask Rate=0.3, Training Protocol=TF, Base Model=Qwen3-4B2026.0564.63
Qwen3-4B + WeMask(SFT)Mask Rate=0.1, Training Protocol=SFT, Base Model=Qwen3-4B2026.0564.54
Qwen3-4B + SFT + WeMask(TF)Mask Rate=0.1, Training Protocol=TF, Base Model=Qwen3-4B2026.0564.24
Qwen3-4B + SFT + WeMask(TF)Mask Rate=0.7, Training Protocol=TF, Base Model=Qwen3-4B2026.0564.22
Qwen3-4B + SFT + WeMask(TF)Mask Rate=0.5, Training Protocol=TF, Base Model=Qwen3-4B2026.0564.19
Qwen3-4B + SFTMask Rate=-, Training Protocol=SFT, Base Model=Qwen3-4B2026.0564.15
Qwen3-4B + WeMask(SFT)Mask Rate=0.3, Training Protocol=SFT, Base Model=Qwen3-4B2026.0564.06
Qwen3-4B + WeMask(SFT)Mask Rate=0.5, Training Protocol=SFT, Base Model=Qwen3-4B2026.0563.76
Qwen3-4B + WeMask(SFT)Mask Rate=0.7, Training Protocol=SFT, Base Model=Qwen3-4B2026.0563.62
Qwen3-4B + WeMask(SFT)Mask Rate=1.0, Training Protocol=SFT, Base Model=Qwen3-4B2026.0562.05
Qwen3-4B + SFT + WeMask(TF)Mask Rate=1.0, Training Protocol=TF, Base Model=Qwen3-4B2026.0561.75