Scaling Personality Control in LLMs with Big Five Scaler Prompts

Gunhee Cho; Yun-Gyung Cheong

cs.CLAug 8, 2025

Scaling Personality Control in LLMs with Big Five Scaler Prompts

Authors: Gunhee Cho, Yun-Gyung Cheong

Organizations: Sungkyunkwan University, South Korea

Abstract

We present Big5-Scaler, a prompt-based framework for conditioning large language models (LLMs) with controllable Big Five personality traits. By embedding numeric trait values into natural language prompts, our method enables fine-grained personality control without additional training. We evaluate Big5-Scaler across trait expression, dialogue generation, and human trait imitation tasks. Results show that it induces consistent and distinguishable personality traits across models, with performance varying by prompt type and scale. Our analysis highlights the effectiveness of concise prompts and lower trait intensities, providing a efficient approach for building personality-aware dialogue agents.

Figures & tables

Figure 1: Overview of the Big5-Scaler method. Persona prompts are constructed using one of three prompt types (simple, specific, or simspec) and assigned to agents based on Big Five trait values. In this example, the scale n is set to 100. Two agents with different trait configurations engage in dialogue, during which their utterances are generated and stored in each agent’s memory.

Agent Model	RMSE
llama3-8b	1.822
mistral-14b	1.785
phi4-25b	1.8

Table 5: RMSE scores of each Big5-Scaler agent model in human personality imitation

Appendix figures & tables2 assets

Supplementary material from the paper’s appendix.

Appendix

Figure 3: Trait Tendency Curves across different models: LLaMA, Mistral, and Phi-4

Model	Scale	Prompt	BFI	IPIP-NEO	Average
llama3-8b	10	simple	27.715	34.301	31.008
		specific	24.638	34.584	29.611
		simspec	24.675	34.944	29.810
	25	simple	30.331	33.347	31.839
		specific	28.866	34.820	31.843
		simspec	30.094	35.049	32.572

Appendix

Table 7: RMSE scores across different Big Five trait questionnaires (BFI, IPIP-NEO) using various combinations of models, scaling levels, and prompting methods. Bold values indicate the best-performing prompt within each model setting for each questionnaire, while red values represent the best overall performance per questionnaire across all model settings. Overall, the configuration using phi4-14b with a scale of 10 and the simple prompt achieved the best performance.

Persona-Conditioned Generation

Persona-conditioned generation uses profiles, traits, or histories to steer AI outputs toward specified identities, preferences, and behaviors.

See all 102 papers

Personality, Role, and Expressive Style in Large Language Models: An Interactionist Analysis
May 27, 2026Moe Nagao, Koichiro Terao, Mikio Nakano +1Large Language Model-Based Role-Play Simulation LLM Prompting
Okayama Prefectural University · AI & Humans Lab · C4A Research Institute
PERSONA: Dynamic and Compositional Inference-Time Personality Control via Activation Vector Algebra
Feb 17, 2026Xiachong Feng, Liang Zhao, Weihong Zhong +5Controllable Text Generation Personality Modeling in Language Models
Harbin Institute of Technology · The University of Hong Kong
Mechanistic Personality Analysis of LLMs Steering Personality via Latent Feature Interventions
Jun 27, 2026David Courtis, Ting HuLanguage Model Steering LLM Interpretability
Department of Computing, Queen’s University Kingston, Kingston, ON, Canada.

May 27, 2026J/K move · Enter open · S save

Personality, Role, and Expressive Style in Large Language Models: An Interactionist Analysis

Moe Nagao, Koichiro Terao, Mikio Nakano, Naoto Iwahashi

Okayama Prefectural University · AI & Humans Lab · C4A Research Institute

Large Language Model-Based Role-Play Simulation LLM Prompting Personality Modeling in Language Models Controllable Text Generation Persona-Conditioned Generation

Open paper AskPDF arXiv

Prompt-based personality control is a key technique for designing large language model (LLM) dialogue agents that behave consistently across social contexts. However, specifying Big Five personality traits (BFTs) in a prompt does not ensure that the intended traits are expressed in generated utterances. This paper investigates this mismatch from an interactionist perspective, viewing personality expression as a context-dependent outcome shaped by the interplay between trait specification and situational factors. We analyze how perceived BFT expression in LLM-generated dialogue is influenced by three prompt factors: personality traits, dialogue roles, and expressive styles. Using a factorial design that combines six personality conditions, three roles, and three expressive-style conditions, we generate 1,080 LLM-agent dialogues in each of English and Japanese. We then evaluate the target agent's utterances using an LLM-as-a-judge framework to estimate expressed Big Five traits. The results show that expressed personality is shaped not only by explicit trait specification, but also by dialogue role and expressive style. These effects are trait-specific: dialogue role strongly influences Openness, expressive style substantially shapes Conscientiousness and Agreeableness, and explicit trait specification dominates Neuroticism. Even without explicit personality-trait specification, social and expressive conditions induce distinct personality-like impressions. Cross-linguistic comparisons show broadly similar patterns between English and Japanese dialogues, with noticeable differences only under specific combinations of personality, role, and expressive style. These findings suggest that personality control in LLM agents should be understood not as a direct consequence of trait prompting, but as a context-dependent process involving personality specification, social role, and expressive style.

Abstract

Persona-Conditioned Generation

Explore similar work

Personality, Role, and Expressive Style in Large Language Models: An Interactionist Analysis

PERSONA: Dynamic and Compositional Inference-Time Personality Control via Activation Vector Algebra

Mechanistic Personality Analysis of LLMs Steering Personality via Latent Feature Interventions