AINeutralarXiv – CS AI · Apr 106/10
🧠
Restoring Heterogeneity in LLM-based Social Simulation: An Audience Segmentation Approach
Researchers demonstrate that Large Language Models used for social simulation produce more accurate behavioral predictions when trained with audience segmentation strategies rather than averaged personas. The study finds that moderate identifier granularity and data-driven selection methods optimize structural and predictive fidelity, with no single configuration excelling across all evaluation dimensions.
🧠 Llama