Personality Modeling in Language Models

Overview

This field asks whether language models show stable personality-like tendencies, how those tendencies relate to human traits, and whether models can simulate individuals or populations. Researchers measure them with Big Five and other psychometric questionnaires, human-grounded behavioral tasks, interviews, and interaction histories; they also use prompting, fine-tuning, memory, and activation steering to shape persona expression. Benchmarks increasingly test behavior across situations and multiple turns, not just self-reports or fluent role-play. A central tension is that questionnaire scores and apparent persona fidelity may reflect prompt artifacts or response biases rather than behavior, while stronger steering can affect capabilities, safety, and consistency.

Latest papers

CardsList

185 papers15 papers in the last four weeks, down 12% on the four weeks before. 0.1% of all new papers.

  1. Human Psychometric Questionnaires Mischaracterize LLM Behavior

    Sep 12, 2025Woojung Song, Dongmin Choi, Yoonah Park +3Personality Modeling in Language ModelsPsychometric Validation

  2. PersonaFuse: A Situation-Aware Post-Training Framework for Adaptive LLM Behavior in Enterprise Tasks

    Sep 9, 2025Yixuan Tang, Yi Yang, Ahmed AbbasiLanguage Model Post-TrainingPersonality Modeling in Language Models

  3. Localizing Persona Representations in LLMs

    May 30, 2025Celia Cintas, Miriam Rateike, Erik Miehling +2LLM InterpretabilityPersonality Modeling in Language Models

  4. AI Simulations Grounded in Rich Self-Reports Increase Accuracy and Reduce Distortion

    Nov 15, 2024Joon Sung Park, Carolyn Q. Zou, Jonne Kamphorst +8Personality Modeling in Language ModelsHuman Behavior Simulation