List your company

Translator-call mode selection on PolyMath Medium

75.22Macro F1LUAR
37.790447.507757.22566.9423Jun 1, 2026

Evaluation results

MethodLinks
LUARBackbone=Qwen3-4B2026.0675.22
ST(qr)Backbone=Qwen3-8B2026.0669.19
BOUNDARY-SFTBackbone=Qwen3-8B2026.0665.9
LUARBackbone=Qwen3-8B2026.0665.61
ST(qr)Backbone=Qwen3-4B2026.0664.59
BOUNDARY-SFTBackbone=Qwen3-4B2026.0662.22
SELF-ASSESSMENTBackbone=Qwen3-4B2026.0656.44
ST(q)Backbone=Qwen3-8B2026.0656.18
SELF-ASSESSMENTBackbone=Qwen3-8B2026.0654.28
ST(q)Backbone=Qwen3-4B2026.0652.59
NATIVE-TOOL-USEBackbone=Qwen3-4B2026.0649.61
NATIVE-TOOL-USEBackbone=Qwen3-8B2026.0639.23