List your company

Multi-turn role-play on MSD

83.2Success Rate (SR)EmoLLM
77.37678.88880.481.912Mar 17, 2026

Evaluation results

MethodLinks
EmoLLMBackbone=Qwen3-8B, Method Category=Ours2026.0383.22.864.173.71
PERMBackbone=Qwen3-8B, Method Category=RL Methods2026.0382.12.744.23.64
ICL-AIFBackbone=Qwen3-8B, Method Category=Prompting Methods2026.0381.52.814.163.62
SFTBackbone=Qwen3-8B, Method Category=SFT Methods2026.0381.22.743.973.37
RLVERBackbone=Qwen3-8B, Method Category=RL Methods2026.0381.22.884.183.61
ARG-PromptBackbone=Qwen3-8B, Method Category=Prompting Methods2026.03812.884.143.58
RLFF-ESCBackbone=Qwen3-8B, Method Category=RL Methods2026.0380.62.964.153.55
PPDPBackbone=Qwen3-8B, Method Category=SFT Methods2026.0380.42.934.113.54
Best-of-NBackbone=Qwen3-8B, Method Category=Prompting Methods2026.0379.63.124.083.51
ESCoTBackbone=Qwen3-8B, Method Category=SFT Methods2026.0379.43.064.183.47
Ask-an-ExpertBackbone=Qwen3-8B, Method Category=Prompting Methods2026.0379.13.154.053.46
Qwen3-8BBackbone=Qwen3-8B, Method Category=Base Model2026.0377.63.274.133.41