arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2503.04735cs.CYcs.LG
收藏

人格特质如何塑造大语言模型的风险承担行为

How Personality Traits Shape LLM Risk-Taking Behaviour

NatWest AI Research(国民西敏寺银行AI研究院);University College London(伦敦大学学院)

John Hartley, Conor Hamill, Devesh Batra, Dale Seddon, Ramin Okhrati, Raad Khraishi

AI总结:

本研究结合累积前景理论和大五人格框架,发现GPT-4o在风险选择中呈风险中性,且开放性人格干预对其风险倾向影响最大,为塑造大语言模型决策行为提供了依据。

摘要:

大语言模型(LLMs)正越来越多地被部署为自主智能体,这要求我们更深入地理解其在风险条件下的决策行为。本研究采用累积前景理论(CPT)和大五人格框架,考察大语言模型的人格特质与风险倾向之间的关系。我们聚焦于GPT-4o,将其行为与人类基线及早期模型进行比较。研究发现,与人类平均水平相比,GPT-4o表现出更高的尽责性和宜人性特质,同时在选项选择中表现为风险中性的理性智能体。对GPT-4o的大五人格特质进行干预,尤其是开放性,会显著影响其风险倾向,这与人类研究中观察到的模式一致。值得注意的是,开放性是GPT-4o风险倾向中最具影响力的因素,与人类研究结果相符。相比之下,GPT-4-Turbo等旧模型在人格—风险关系的泛化上表现不一致。本研究增进了我们对大语言模型在风险条件下行为的理解,并阐明了基于人格的干预在塑造大语言模型决策方面的潜力与局限。我们的研究结果对开发更稳健、更可预测的人工智能系统(如金融建模)具有重要意义。

英文摘要:

Large Language Models (LLMs) are increasingly deployed as autonomous agents, necessitating a deeper understanding of their decision-making behaviour under risk. This study investigates the relationship between LLMs' personality traits and risk propensity, employing cumulative prospect theory (CPT) and the Big Five personality framework. We focus on GPT-4o, comparing its behaviour to human baselines and earlier models. Our findings reveal that GPT-4o exhibits higher Conscientiousness and Agreeableness traits compared to human averages, while functioning as a risk-neutral rational agent in prospect selection. Interventions on GPT-4o's Big Five traits, particularly Openness, significantly influence its risk propensity, mirroring patterns observed in human studies. Notably, Openness emerges as the most influential factor in GPT-4o's risk propensity, aligning with human findings. In contrast, legacy models like GPT-4-Turbo demonstrate inconsistent generalization of the personality-risk relationship. This research advances our understanding of LLM behaviour under risk and elucidates the potential and limitations of personality-based interventions in shaping LLM decision-making. Our findings have implications for the development of more robust and predictable AI systems such as financial modelling.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑