List your company

Single Product shopping assistance on SHOPPING COMPANION benchmark

90AccuracyQwen3-4B-LoRA + RL (Dual&Tool-wise Reward)
47.3658.4369.580.57Mar 16, 2026

Evaluation results

MethodLinks
Qwen3-4B-LoRA + RL (Dual&Tool-wise Reward)Category=Ours2026.039084
Qwen3-4B-LoRA + RL (Dual-reward)Category=Ours2026.038981
GPT-4.1Category=Closed2026.038878
GPT-5Category=Closed2026.038275
Qwen3-4B-LoRACategory=Ours2026.038272
Qwen3-MaxCategory=Closed2026.038072
GPT-4oCategory=Closed2026.037972
Qwen3-Next-80B-A3BCategory=Open2026.036357
Qwen3-30B-A3BCategory=Open2026.036053
Qwen3-4BCategory=Open2026.034944