Social Choice Foundations for Simulation-Augmented Generation

2026-09-29

AI Summary

This study addresses the inherent trade-off between inference efficiency and population representativeness in simulated augmented generation, providing its first formal definition. To reconcile these objectives, this work proposes a dynamic routing framework grounded in social choice theory. Methodologically, it introduces the mPJR+ axiom and proves that overall proportional representativeness can be guaranteed by routing only a small subset of simulated individuals. By integrating proportional clustering theory with large language model-based simulation techniques, the approach enables highly efficient generation. Experimental evaluations on political decision-making and personalized advice tasks demonstrate that the proposed algorithm achieves significantly higher mPJR+ satisfaction than K-means and random baselines. Ultimately, this framework facilitates high-quality representative generation with minimal computational overhead.

Abstract

Simulation-augmented generation (SAGE) is a recent technical proposal in which models simulate individuals' viewpoints at inference time in order to provide more representative answers to contentious user queries. A core challenge for SAGE is making inference-time simulation efficient without sacrificing representation quality. We introduce the first formalization of this problem, based upon an axiom from proportional clustering known as metric proportional justified representation+ (mPJR+) which is the strongest proportionality axiom known to always be satisfiable by centroid-based clustering. We prove that to proportionally represent the viewpoints of a population of $n_H$ humans on a given prompt, we need only create simulations of $n \ll n_H$ individuals, and at inference time, need only dynamically route to $k \ll n$ of those simulations based upon the prompt. This twofold reduction still yields approximate proportional representation guarantees for the entire population. Empirically, across two domains-political questions and personal advice-our proposed routing algorithm achieves higher mPJR+ satisfaction rates than $k$-means-based or random selection baselines.

Problem

• Simulation-Augmented Generation
• Social Choice
• Proportional Representation
• Inference Efficiency

Innovation

• Simulation-Augmented Generation
• Proportional Representation
• Metric Proportional Justified Representation+
• Dynamic Routing
• Social Choice Theory