Building Social World Models with Large Language Models

2026-06-09

AI Summary

This study addresses the modeling and prediction of dynamic social belief evolution under the influence of major events. To this end, it proposes the Social World Model (SWM) framework, which, for the first time, leverages large language models to learn an unsupervised state transition function for social beliefs without requiring human annotations or survey data. The approach integrates variational inference with temporal pattern mining, optimizing the evidence lower bound to yield an interpretable model of belief dynamics. Evaluated on a newly curated SWM-bench benchmark, the method significantly outperforms existing temporal foundation models, achieves state-of-the-art performance on the Kalshi dataset, and demonstrates competitive results on Polymarket, thereby validating its effectiveness and generalization capability.

Abstract

Understanding and predicting how social beliefs evolve in response to events -- from policy changes to scientific breakthroughs -- remains a fundamental challenge in social science. Given LLMs' commonsense knowledge and social intelligence, we ask: Can LLMs model the dynamics of social beliefs following social events? In this work, we introduce the concept of the Social World Model (SWM), a general framework designed to capture how social beliefs evolve in response to major events. SWM learns state-transition functions for social beliefs by mining temporal patterns in social data and optimizing the evidence lower bound, without the need for explicit human annotations linking events to belief shifts, or for expensive census data. To evaluate SWM, we introduce a benchmark, SWM-bench, derived from real-world prediction markets, specifically Kalshi and Polymarket. SWM-bench includes over 12k data points for social belief prediction tasks spanning diverse domains such as politics, finance, and cryptocurrency. Our experimental results show that SWM significantly outperforms time-series foundation models, achieving state-of-the-art results on Kalshi data and demonstrating competitive performance on Polymarket data, while offering interpretable insights into the underlying mechanisms of social belief dynamics.

Problem

• social belief dynamics
• social events
• belief evolution
• prediction markets
• social world modeling

Innovation

• Social World Model
• Large Language Models
• Belief Dynamics
• Prediction Markets
• Unsupervised Learning