The Epi-LLM Framework: probing LLM behavioral priors through epidemiological agent-based models
arXiv:2606.02867v1 Announce Type: cross Abstract: Human behaviour during epidemics affects infectious disease dynamics, but quantifying this remains deeply challenging. Here we introduce the Epi-LLM framework: a novel integration of agent-based modelling, real-life epigames, and large language models (LLMs) in which a synthetic society of agents reasons and adapts dynamically over an outbreak contact network. Comparing synthetic agent behaviour against a no-intervention SEIR baseline and human participant data from the AUIB epigame study, we find that LLM agents across four different architect
The proliferation of advanced large language models allows for their application in complex agent-based simulations, reflecting a growing trend in using AI for social and epidemiological modeling.
This development offers a novel, scalable approach to understanding and predicting human behavior during health crises, which is crucial for effective public health interventions and policy.
LLMs can now be integrated into sophisticated epidemiological models, providing a more dynamic and nuanced simulation of societal responses to outbreaks compared to traditional, static models.
- · Public health researchers
- · Epidemiologists
- · AI model developers
- · Policy makers
- · Traditional static modeling approaches
- · Researchers reliant solely on costly human trials for behavioral insights
LLMs provide simulated behavioral data that improves the accuracy of epidemic prediction models.
Enhanced predictive capabilities lead to more effective, AI-informed public health strategies and resource allocation during future crises.
The success of LLMs in epidemiological modeling could spur their application in other complex social science simulations, transforming fields like economics, sociology, and political science.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI