SocialFiVis: A Visual Analytics Sandbox for LLM-Grounded Multi-Agent Simulation in Social Finance

2026-08-09

AI Summary

This study addresses the complex nonlinear interactions among governance rules, individual behaviors, and economic outcomes in Social Finance (SocialFi), which existing methods struggle to simulate, particularly regarding counterfactual intervention effects. Building upon the Institutional Analysis and Development (IAD) framework, this work proposes the first IAD-theory-informed visual analytics system integrated with a large language model (LLM)-driven multi-agent simulation sandbox. The system incorporates a Perception-Reasoning-Action (PRA) operational mechanism, featuring LLM-generated personalized agents, mechanism-guided runtime logic, hierarchical multi-view interfaces, and interpretable reasoning pathways, enabling traceable analysis from individual actions to emergent system-level phenomena. Case studies and user experiments demonstrate that the approach effectively supports fine-grained behavioral attribution, reveals emergent patterns such as the structural decoupling between social capital and financial health, and elucidates the emergence of community resilience under localized governance shocks.

Abstract

The emergence of social finance (SocialFi) transforms online communities into complex socio-economic systems. Within these spaces, collective decisions shape a "digital commons" characterized by social capital (e.g., community trust) and financial health (e.g., market liquidity). Governing such hybrid ecosystems is challenging because real-world interventions are costly and irreversible. While counterfactual simulation is essential for exploring alternative governance strategies, existing approaches fail to capture the non-linear interplay between governance rules, individual behaviors, and emergent economic outcomes. To systematically unpack this complexity, we operationalize the Institutional Analysis and Development (IAD) framework as our theoretical foundation, synthesizing prior literature with insights from formative expert interviews. Built on this framework, we present SocialFiVis, an IAD-embedded visual analytics sandbox. It introduces a robust model to quantify the dual-track digital commons, coupled with a two-phase simulation engine. This engine combines LLM-derived personas with a mechanism-guided Perception-Reasoning-Action (PRA) runtime to simulate heterogeneous, context-aware agents empirically grounded in the retained messaging cohort. A hierarchical multi-view interface with interpretable reasoning pathways enables community operators to explore counterfactual policies and trace system-level outcomes back to individual behavioral rationales. We evaluate SocialFiVis through two case studies, a user study, and follow-up interviews. Results demonstrate that SocialFiVis supports fine-grained behavioral attribution and helps explain emergent phenomena such as the structural decoupling of social capital and the resilience of messaging members under localized governance shocks.

Problem

• Social Finance
• Governance
• Counterfactual Simulation
• Digital Commons
• Agent Behavior

Innovation

• Visual Analytics
• LLM-Grounded Multi-Agent Simulation
• Institutional Analysis and Development (IAD)
• Perception-Reasoning-Action (PRA) Framework
• Social Finance