List your company

Multi-turn role-play on MedD

95.3Success Rate (SR)EmoLLM
88.95690.60392.2593.897Mar 17, 2026

Evaluation results

MethodLinks
EmoLLMBackbone=Qwen3-8B, Method Category=Ours2026.0395.32.184.084.59
PERMBackbone=Qwen3-8B, Method Category=RL Methods2026.03942.284.14.18
RLVERBackbone=Qwen3-8B, Method Category=RL Methods2026.03932.364.084.01
ICL-AIFBackbone=Qwen3-8B, Method Category=Prompting Methods2026.0392.72.314.064.06
RLFF-ESCBackbone=Qwen3-8B, Method Category=RL Methods2026.0392.42.444.023.92
ARG-PromptBackbone=Qwen3-8B, Method Category=Prompting Methods2026.0392.32.284.024.01
PPDPBackbone=Qwen3-8B, Method Category=SFT Methods2026.0391.82.3743.89
ESCoTBackbone=Qwen3-8B, Method Category=SFT Methods2026.0391.2—4.013.76
Best-of-NBackbone=Qwen3-8B, Method Category=Prompting Methods2026.0390.82.413.923.97
Ask-an-ExpertBackbone=Qwen3-8B, Method Category=Prompting Methods2026.0390.82.633.963.72
SFTBackbone=Qwen3-8B, Method Category=SFT Methods2026.0390.42.293.913.68
Qwen3-8BBackbone=Qwen3-8B, Method Category=Base Model2026.0389.22.923.953.57