List your company

Beauty Live-Commerce Dialogue Generation on beauty live-commerce evaluation set

4.32InformativenessVerbalValue
2.4482.9343.423.906May 14, 2026

Evaluation results

MethodLinks
VerbalValue2026.054.324.914.624.45864.524.61
Claude Sonnet 4.62026.053.514.434.914.27554.234.55
GPT-5.42026.053.464.854.864.21733.953.87
Gemini 3.1 Pro2026.052.883.763.783.18383.243.13
Qwen2.5-32B-InstructModel size=32B, Variant=Instruct2026.052.523.212.252.36442.582.84