arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2502.08599cs.CL
收藏

SPeCtrum:基于LLM智能体的多维度身份表示基础框架

SPeCtrum: A Grounded Framework for Multidimensional Identity Representation in LLM-Based Agent

Seoul National University(首尔国立大学)

Keyeun Lee, Seo Hyeong Kim, Seolhee Lee, Jinsu Eun, Yena Ko, Hayeon Jeon, Esther Hehsun Kim, Seonghye Cho, Soeun Yang, Eun-mee Kim, Hajin Lim

AI总结:

针对现有方法过度简化人类复杂性导致身份表示不完整的问题,提出多维度身份表示框架SPeCtrum,通过整合社会身份、个人身份和个人生活背景构建真实LLM智能体角色,经评估证实其能增强身份表示的真实性和准确性。

摘要:

现有模拟个体身份的方法往往过度简化人类复杂性,可能导致不完整或扁平化的表示。为此,我们提出SPeCtrum,这是一个通过整合个体多维度自我概念来构建真实LLM智能体角色的基础框架。SPeCtrum集成了三个核心组件:社会身份(S)、个人身份(P)和个人生活背景(C),每个组件贡献身份中独特但相互关联的方面。为评估SPeCtrum在身份表示中的有效性,我们进行了自动化和人类评估。使用流行戏剧角色的自动化评估显示,从偏好和日常例行短文中提取的个人生活背景(C)比单独的社会身份(S)和个人身份(P)更有效地建模角色身份,且表现与完整SPC组合相当。相比之下,涉及现实世界个体的人类评估发现,完整SPC组合比单独的C提供了更全面的自我概念表示。我们的研究结果表明,虽然单独的C可能足以进行基本身份模拟,但整合S、P和C可增强现实世界身份表示的真实性和准确性。总体而言,SPeCtrum为LLM智能体中的个体模拟提供了结构化方法,实现了更个性化的人机交互,并提高了基于模拟的行为研究的真实性。

英文摘要:

Existing methods for simulating individual identities often oversimplify human complexity, which may lead to incomplete or flattened representations. To address this, we introduce SPeCtrum, a grounded framework for constructing authentic LLM agent personas by incorporating an individual's multidimensional self-concept. SPeCtrum integrates three core components: Social Identity (S), Personal Identity (P), and Personal Life Context (C), each contributing distinct yet interconnected aspects of identity. To evaluate SPeCtrum's effectiveness in identity representation, we conducted automated and human evaluations. Automated evaluations using popular drama characters showed that Personal Life Context (C)-derived from short essays on preferences and daily routines-modeled characters' identities more effectively than Social Identity (S) and Personal Identity (P) alone and performed comparably to the full SPC combination. In contrast, human evaluations involving real-world individuals found that the full SPC combination provided a more comprehensive self-concept representation than C alone. Our findings suggest that while C alone may suffice for basic identity simulation, integrating S, P, and C enhances the authenticity and accuracy of real-world identity representation. Overall, SPeCtrum offers a structured approach for simulating individuals in LLM agents, enabling more personalized human-AI interactions and improving the realism of simulation-based behavioral studies.

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。

↑