Think-Before-Speak framework for multi-agent social simulation

Arxiv CS.AI (top papers)

Proposes a method where agents internally evaluate before public expression in LLM-based multi-agent simulations. Aims to improve deliberation dynamics and opinion formation in social simulations.

multi-agent-systemssocial-simulationllm

Arxiv CS.AI (top papers)·Kaiqi YangTai-Quan PengSanguk LeeHui Liu

8 days ago