Language Agents as Digital Representatives in Collective Decision-Making

2025-02-13

AI Summary

This paper formulates the “Language Agent as Human Digital Representative” problem, aiming to enable large language models (LLMs) to faithfully encode and express the true preferences of represented individuals in collective decision-making. Methodologically, it frames digital representation as a behavior optimization problem subject to mechanism output equivalence constraints; it fine-tunes LLMs using human preference data and develops an experimental framework integrating consensus protocols and group decision mechanisms. The key contribution is the first formal, mechanism-design–informed definition of verifiability criteria for digital representatives. Empirically, preference-aware fine-tuned LLMs consistently reproduce the decision distributions observed across diverse human participants in consensus tasks—preserving individual representativeness while significantly improving both group decision efficiency and fairness.

Technology Category

• Knowledge Representation and Reasoning: Preferences
• Humans and AI: Learning Human Values and Preferences
• Multiagent Systems: Mechanism Design

Application Category

• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Consider the process of collective decision-making, in which a group of individuals interactively select a preferred outcome from among a universe of alternatives. In this context,"representation"is the activity of making an individual's preferences present in the process via participation by a proxy agent -- i.e. their"representative". To this end, learned models of human behavior have the potential to fill this role, with practical implications for multi-agent scenario studies and mechanism design. In this work, we investigate the possibility of training extit{language agents} to behave in the capacity of representatives of human agents, appropriately expressing the preferences of those individuals whom they stand for. First, we formalize the setting of extit{collective decision-making} -- as the episodic process of interaction between a group of agents and a decision mechanism. On this basis, we then formalize the problem of extit{digital representation} -- as the simulation of an agent's behavior to yield equivalent outcomes from the mechanism. Finally, we conduct an empirical case study in the setting of extit{consensus-finding} among diverse humans, and demonstrate the feasibility of fine-tuning large language models to act as digital representatives.

Problem

• Training language agents as human representatives
• Simulating agent behavior for decision outcomes
• Fine-tuning models for consensus-finding scenarios

Innovation

• Language agents simulate human preferences
• Fine-tuned large language models represent humans
• Digital representation in collective decision-making