Human Simulacra: Benchmarking the Personification of Large Language Models

TL;DR

This paper presents a framework for creating and evaluating human-like virtual characters using large language models, aiming to simulate human cognition and personality for research and practical applications.

Contribution

It introduces a novel personification framework, including story construction, multi-agent cognitive simulation, and psychology-guided evaluation methods for LLMs.

Findings

Simulacra responses align with target characters

Framework effectively models human cognitive processes

Potential for reducing research costs in social science

Abstract

Large language models (LLMs) are recognized as systems that closely mimic aspects of human intelligence. This capability has attracted attention from the social science community, who see the potential in leveraging LLMs to replace human participants in experiments, thereby reducing research costs and complexity. In this paper, we introduce a framework for large language models personification, including a strategy for constructing virtual characters' life stories from the ground up, a Multi-Agent Cognitive Mechanism capable of simulating human cognitive processes, and a psychology-guided evaluation method to assess human simulations from both self and observational perspectives. Experimental results demonstrate that our constructed simulacra can produce personified responses that align with their target characters. Our work is a preliminary exploration which offers great potential in practical applications. All the code and datasets will be released, with the hope of inspiring further investigations. Our code and dataset are available at: https://github.com/hasakiXie123/Human-Simulacra.