Modeling U.S. Attitudes Toward China via an Event-Steered Multi-Agent Simulator
Researchers have developed an Event-Steered Multi-Agent Simulator (ES-MAS) to model the dynamic evolution of public opinion, specifically U.S. attitudes toward China. This new simulator addresses limitations in existing models by incorporating significant real-world events and daily news to drive agent interactions and opinion shifts. The system utilizes a novel China-U.S. Relation Evolution (CURE) dataset and a Dual-Stream Data Integration Engine to align simulations with historical timelines and personalize information exposure. AI
IMPACT Provides a more realistic framework for modeling geopolitical opinion dynamics, potentially aiding risk assessment.