List your company

Mathematical Reasoning on Minerva Math (Avg@8)

23.9Avg@8EAD
6.448810.979415.5120.0406Feb 13, 2026

Evaluation results

MethodLinks
EADBackbone=Qwen3-4B-Base, Schedule=1.2→0.12026.0223.9
IntroLLMBackbone=Qwen3-4B-Base2026.0223.48
TAMPOBackbone=Qwen3-4B-Base, Range=0.6-1.52026.0222.79
GRPOBackbone=Qwen3-4B-Base, Temperature (τ)=1.22026.0222.61
GRPOBackbone=Qwen3-4B-Base, Temperature (τ)=1.02026.0221.51
GRPOBackbone=Qwen3-4B-Base, Temperature (τ)=0.62026.0221.09
IntroLLMBackbone=Qwen3-1.7B-Base2026.0218.29
GRPOBackbone=Qwen3-1.7B-Base, Temperature (τ)=1.22026.0217.33
EADBackbone=Qwen3-1.7B-Base, Schedule=1.2→0.12026.0217.33
GRPOBackbone=Qwen3-1.7B-Base, Temperature (τ)=1.02026.0217.28
TAMPOBackbone=Qwen3-1.7B-Base, Range=0.6-1.52026.0216.64
GRPOBackbone=Qwen3-1.7B-Base, Temperature (τ)=0.62026.0215.17
BaseBackbone=Qwen3-4B-Base2026.0212.5
BaseBackbone=Qwen3-1.7B-Base2026.027.12