Context-Value-Action Architecture for Value-Driven Large Language Model Agents

TL;DR

This paper introduces the Context-Value-Action (CVA) architecture for LLM agents, which reduces behavioral rigidity and value polarization by explicitly modeling dynamic human values, outperforming existing methods.

Contribution

The paper proposes the CVA architecture, grounded in psychological theories, with a novel Value Verifier trained on real data to improve LLM agent behavior and interpretability.

Findings

CVA outperforms baseline methods on CVABench with over 1.1 million interaction traces.

CVA reduces value polarization and enhances behavioral fidelity.

CVA improves interpretability of LLM agent decisions.

Abstract

Large Language Models (LLMs) have shown promise in simulating human behavior, yet existing agents often exhibit behavioral rigidity, a flaw frequently masked by the self-referential bias of current "LLM-as-a-judge" evaluations. By evaluating against empirical ground truth, we reveal a counter-intuitive phenomenon: increasing the intensity of prompt-driven reasoning does not enhance fidelity but rather exacerbates value polarization, collapsing population diversity. To address this, we propose the Context-Value-Action (CVA) architecture, grounded in the Stimulus-Organism-Response (S-O-R) model and Schwartz's Theory of Basic Human Values. Unlike methods relying on self-verification, CVA decouples action generation from cognitive reasoning via a novel Value Verifier trained on authentic human data to explicitly model dynamic value activation. Experiments on CVABench, which comprises over 1.1 million real-world interaction traces, demonstrate that CVA significantly outperforms baselines. Our approach effectively mitigates polarization while offering superior behavioral fidelity and interpretability.