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

2025-02-12

AI Summary

Existing identity modeling approaches for LLM agents oversimplify the complexity of human selfhood, resulting in flattened, inadequate representations. Method: This paper proposes a novel three-dimensional decoupled–coupled identity modeling paradigm grounded in self-concept theory, systematically integrating Social identity (S), Personal identity (P), and Contextual circumstances (C) to construct a multidimensional, individual-centric identity framework. Our approach combines prompt engineering, role-embedding fine-tuning, and multi-granularity evaluation (automated + human), leveraging short-text generation and behavioral consistency discrimination. Contribution/Results: Experiments demonstrate that, in dramatic role simulation, the Contextual (C) dimension alone achieves performance comparable to the full S+P+C model; in real-person modeling, joint SPC representation significantly enhances both authenticity and comprehensiveness. This work establishes an interpretable, scalable theoretical and practical paradigm for identity modeling in LLM agents.

Technology Category

• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction
• Machine Learning: Large Multimodal Models (LMMs)
• Multiagent Systems: Modeling other Agents

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Agentic search

Abstract

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.

Problem

• Simulates multidimensional human identities in AI
• Enhances authenticity of LLM agent personas
• Improves realism in human-AI interactions

Innovation

• Multidimensional identity representation framework
• Integrates Social, Personal, Life Context
• Enhances authenticity in LLM agents