List your company

E-commerce Task Evaluation on Classic-B n=5,091 (held-out)

84.8B1GPT-5.5
84.17684.33884.584.662Jul 8, 2026

Evaluation results

MethodLinks
GPT-5.52026.0784.875.557.972.872.7
Our LBMtraining_stage=base + SFT-Lotus2026.0784.297.671.17682.2