List your company

Dense Model Loss Curve Prediction WSD to Cosine transfer

0.202MAPECMMT (λ = 0.99)
0.120040.673271.22651.77973Dec 5, 2025

Evaluation results

MethodLinks
CMMT (λ = 0.99)Model Size=100M2025.120.202
ASMTModel Size=100M2025.120.232
CMMT (λ = 0.99)Model Size=500M2025.120.379
CMMT (λ = 0.999)Model Size=500M2025.120.417
ASMTModel Size=500M2025.120.453
ASMTModel Size=50M2025.120.456
CMMT (λ = 0.99)Model Size=50M2025.120.539
CMMT (λ = 0.999)Model Size=50M2025.121.886
CMMT (λ = 0.999)Model Size=100M2025.122.251