IntervenSim: Intervention-Aware Social Network Simulation for Opinion Dynamics

2026-04-07

AI Summary

This work addresses a critical limitation in existing large language model–driven social network simulations, which typically assume fixed event trajectories and neglect the dynamic interplay between source-level interventions and collective user interactions in shaping opinion evolution. To overcome this, we propose the first intervention-aware, closed-loop multi-agent simulation framework that explicitly models intervention behaviors through source agents, simulates collective responses via group agents, and integrates an intervention–interaction–feedback mechanism to enable co-evolution. By unifying source interventions and group feedback within a dynamic closed loop, our approach effectively captures secondary virality bursts and attitude shifts triggered by interventions. Evaluated on multiple real-world events, the framework reduces MAPE by 41.6% and DTW error by 66.9%, achieving significantly lower computational overhead with fewer agents.

Abstract

LLM-based social network simulation introduces a new computational approach for modeling event evolution in complex online environments. However, existing methods typically simulate social processes under a fixed event trajectory, treating the event as static once initialized and overlooking intervention dynamics, and thus fail to capture the intrinsic evolution of real social network events, where source-side interventions and collective interactions continuously reshape event trajectories, sometimes leading to secondary popularity explosions and collective attitude shifts. To address this limitation, we introduce an intervention-aware simulation framework, IntervenSim, that models event evolution and intervention in a closed loop. We model event developments and source-side interventions using source agents, and collective crowd reactions using crowd agents, capturing their continuous co-evolution through an intervention-aware mechanism that couples source-side intervention, group interaction, and feedback-driven adjustment of subsequent interventions. Experiments on diverse real-world events show that IntervenSim improves MAPE by 41.6% and DTW by 66.9% over prior frameworks, while reducing computational cost with fewer yet more capable agents. These improvements indicate that IntervenSim not only simulates regular event trajectories more faithfully, but also better captures opinion dynamics under intervention in complex cases.

Problem

• opinion dynamics
• social network simulation
• intervention dynamics
• event evolution
• collective interaction

Innovation

• intervention-aware simulation
• opinion dynamics
• LLM-based social simulation
• closed-loop event evolution
• agent-based modeling