Year: 2026
Type: conference-paper
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 selfverification, 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.CONTEXT 08:00 Woke up early to catch the bus… 10:00 Exhausted, had a large sugary coffee (despite being on a diet ) 12:30 Big burger for lunch, gaining weight again 18:00 Coded so hard I barely moved from my chair
Authors: TianZe Zhang, Sirui Sun, Yuhang Xie, Xin Zhang, Zhiqiang Wu, Guojie Song
Institution: Peking University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7166890108