List your company

Web-based Shopping Simulation on WebShop (test)

34.6Success RateMASA
0.289.1918.127.01May 29, 2026

Evaluation results

MethodLinks
MASABackbone=Qwen3-32B2026.0534.659.97.3
MASABackbone=Qwen3-14B2026.0529.254.48
MASABackbone=Qwen3-8B2026.0528.660.14.7
MASABackbone=Qwen3-4B2026.0526.449.18.4
No SkillBackbone=Qwen3-4B2026.052342.29.5
Base SkillBackbone=Qwen3-4B2026.0519.434.811
DS-AdapterBackbone=Qwen3-4B2026.0519.225.712.4
Base SkillBackbone=Qwen3-32B2026.057.224.212
No SkillBackbone=Qwen3-32B2026.056.635.29.9
Base SkillBackbone=Qwen3-8B2026.05632.69.3
No SkillBackbone=Qwen3-8B2026.054.632.710
DS-AdapterBackbone=Qwen3-8B2026.054.418.212.6
DS-AdapterBackbone=Qwen3-32B2026.053.614.313.3
No SkillBackbone=Qwen3-14B2026.052.819.912.7
DS-AdapterBackbone=Qwen3-14B2026.05212.613.6
Base SkillBackbone=Qwen3-14B2026.051.614.813.5