Parametric Social Identity Injection and Diversification in Public Opinion Simulation

Hugging Face Daily Papers Papers
large-language-models public-opinion-simulation social-diversity representation-learning parametric-injection world-values-survey

Summary

This paper proposes Parametric Social Identity Injection (PSII), a framework that injects parametric representations of demographic attributes into LLM hidden states to improve diversity in public opinion simulation. Experiments on the World Values Survey show it reduces KL divergence and enhances diversity compared to prompt-based methods.

Large language models (LLMs) have recently been adopted as synthetic agents for public opinion simulation, offering a promising alternative to costly and slow human surveys. Despite their scalability, current LLM-based simulation methods fail to capture social diversity, producing flattened inter-group differences and overly homogeneous responses across demographic groups. We identify this limitation as a Diversity Collapse phenomenon in LLM hidden representations, where distinct social identities become increasingly indistinguishable across layers. Motivated by this observation, we propose Parametric Social Identity Injection (PSII), a general framework that injects explicit, parametric representations of demographic attributes and value orientations directly into intermediate hidden states of LLMs. Unlike prompt-based persona conditioning, PSII enables fine-grained and controllable identity modulation at the representation level. Extensive experiments on the World Values Survey using multiple open-source LLMs show that PSII significantly improves distributional fidelity and diversity, reducing KL divergence to real-world survey data while enhancing overall diversity. This work provides new insights into representation-level control of LLM agents and advances scalable, diversity-aware public opinion simulation.
Original Article
View Cached Full Text

Cached at: 06/08/26, 07:14 AM

Paper page - Parametric Social Identity Injection and Diversification in Public Opinion Simulation

Source: https://huggingface.co/papers/2603.16142

Abstract

Large language models suffer from reduced social diversity in public opinion simulation due to identity indistinction in hidden representations, which is addressed through a parametric injection framework that enhances demographic representation fidelity and diversity.

Large language models(LLMs) have recently been adopted as synthetic agents forpublic opinion simulation, offering a promising alternative to costly and slow human surveys. Despite their scalability, current LLM-based simulation methods fail to capture social diversity, producing flattened inter-group differences and overly homogeneous responses across demographic groups. We identify this limitation as aDiversity Collapsephenomenon in LLMhidden representations, where distinct social identities become increasingly indistinguishable across layers. Motivated by this observation, we proposeParametric Social Identity Injection(PSII), a general framework that injects explicit, parametric representations ofdemographic attributesandvalue orientationsdirectly into intermediate hidden states of LLMs. Unlike prompt-based persona conditioning, PSII enables fine-grained and controllable identity modulation at the representation level. Extensive experiments on theWorld Values Surveyusing multiple open-source LLMs show that PSII significantly improves distributional fidelity and diversity, reducingKL divergenceto real-world survey data while enhancing overall diversity. This work provides new insights intorepresentation-level controlof LLM agents and advances scalable, diversity-awarepublic opinion simulation.

View arXiv pageView PDFProject pageGitHub2Add to collection

Get this paper in your agent:

hf papers read 2603\.16142

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper1

#### hexi222/psii-identity-representations Feature Extraction• Updated1 day ago • 1

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2603.16142 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2603.16142 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

More Is Not More: What Matters for Diversity in LLM Opinions?

arXiv cs.CL

A factorial experiment reveals that persona detail does not monotonically increase LLM opinion diversity; interaction architectures explore non-overlapping opinion regions; low-cost interventions like temperature scaling have negligible effects.

Evaluating LLM Simulators as Differentially Private Data Generators

arXiv cs.CL

This paper evaluates LLM-based simulators as generators of differentially private synthetic data, using PersonaLedger to assess whether LLMs can faithfully reproduce statistical distributions from DP-protected personas. While achieving promising fraud detection utility (AUC 0.70 at ε=1), the study identifies significant distribution drift caused by systematic LLM biases that override input statistics.

Beyond Static Personas: Situational Personality Steering for Large Language Models

arXiv cs.CL

This paper introduces IRiS, a training-free framework for situational personality steering in LLMs that moves beyond static persona modeling by identifying and leveraging situation-dependent persona neurons. The approach demonstrates that LLM behavior varies contextually and proposes neuron-based identification, retrieval, and weighted steering methods validated on PersonalityBench and a new SPBench benchmark.