Ask, Answer, and Detect: Role-Playing LLMs for Personality Detection with Question-Conditioned Mixture-of-Experts

• URL: http://arxiv.org/abs/2512.08814v1
• Date: Tue, 09 Dec 2025 17:07:54 GMT
• Title: Ask, Answer, and Detect: Role-Playing LLMs for Personality Detection with Question-Conditioned Mixture-of-Experts
• Authors: Yifan Lyu, Liang Zhang,
• Abstract summary: ROME is a novel framework that explicitly injects psychological knowledge into personality detection.<n>We show that ROME consistently outperforms state-of-the-art baselines in experiments on two real-world datasets.
• Score: 4.618735978506653
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Understanding human personality is crucial for web applications such as personalized recommendation and mental health assessment. Existing studies on personality detection predominantly adopt a "posts -> user vector -> labels" modeling paradigm, which encodes social media posts into user representations for predicting personality labels (e.g., MBTI labels). While recent advances in large language models (LLMs) have improved text encoding capacities, these approaches remain constrained by limited supervision signals due to label scarcity, and under-specified semantic mappings between user language and abstract psychological constructs. We address these challenges by proposing ROME, a novel framework that explicitly injects psychological knowledge into personality detection. Inspired by standardized self-assessment tests, ROME leverages LLMs' role-play capability to simulate user responses to validated psychometric questionnaires. These generated question-level answers transform free-form user posts into interpretable, questionnaire-grounded evidence linking linguistic cues to personality labels, thereby providing rich intermediate supervision to mitigate label scarcity while offering a semantic reasoning chain that guides and simplifies the text-to-personality mapping learning. A question-conditioned Mixture-of-Experts module then jointly routes over post and question representations, learning to answer questionnaire items under explicit supervision. The predicted answers are summarized into an interpretable answer vector and fused with the user representation for final prediction within a multi-task learning framework, where question answering serves as a powerful auxiliary task for personality detection. Extensive experiments on two real-world datasets demonstrate that ROME consistently outperforms state-of-the-art baselines, achieving improvements (15.41% on Kaggle dataset).

Related papers

• From Classification to Ranking: Enhancing LLM Reasoning Capabilities for MBTI Personality Detection [25.825456002235967]
  Personality detection aims to measure an individual's corresponding personality traits through their social media posts.<n>Existing approaches enhance personality trait analysis by leveraging Large Language Models (LLMs)<n>We propose a corresponding reinforcement learning training paradigm for personality detection.
  arXiv  Detail & Related papers  (2026-01-26T15:28:43Z)
• A Computational Framework for Interpretable Text-Based Personality Assessment from Social Media [0.0]
  This thesis presents two datasets -- MBTI9k and PANDORA -- collected from Reddit.<n>The PANDORA dataset contains 17 million comments from over 10,000 users.<n>In response, the SIMPA framework was developed - a computational framework for interpretable personality assessment.
  arXiv  Detail & Related papers  (2025-10-03T08:36:36Z)
• Traits Run Deep: Enhancing Personality Assessment via Psychology-Guided LLM Representations and Multimodal Apparent Behaviors [46.55948528317124]
  We propose a novel personality assessment framework called textittextbfTraits Run Deep.<n>It employs textittextbfpsychology-informed prompts to elicit high-level personality-relevant semantic representations.<n>It devises a textittextbfText-Centric Trait Fusion Network that anchors rich text semantics to align and integrate asynchronous signals from other modalities.
  arXiv  Detail & Related papers  (2025-07-30T04:12:14Z)
• Gensors: Authoring Personalized Visual Sensors with Multimodal Foundation Models and Reasoning [61.17099595835263]
  Gensors is a system that empowers users to define customized sensors supported by the reasoning capabilities of MLLMs.<n>In a user study, participants reported significantly greater sense of control, understanding, and ease of communication when defining sensors using Gensors.
  arXiv  Detail & Related papers  (2025-01-27T01:47:57Z)
• Evaluating Large Language Models with Psychometrics [59.821829073478376]
  This paper offers a comprehensive benchmark for quantifying psychological constructs of Large Language Models (LLMs)<n>Our work identifies five key psychological constructs -- personality, values, emotional intelligence, theory of mind, and self-efficacy -- assessed through a suite of 13 datasets.<n>We uncover significant discrepancies between LLMs' self-reported traits and their response patterns in real-world scenarios, revealing complexities in their behaviors.
  arXiv  Detail & Related papers  (2024-06-25T16:09:08Z)
• LLMvsSmall Model? Large Language Model Based Text Augmentation Enhanced Personality Detection Model [58.887561071010985]
  Personality detection aims to detect one's personality traits underlying in social media posts. Most existing methods learn post features directly by fine-tuning the pre-trained language models. We propose a large language model (LLM) based text augmentation enhanced personality detection model.
  arXiv  Detail & Related papers  (2024-03-12T12:10:18Z)
• Personality Style Recognition via Machine Learning: Identifying Anaclitic and Introjective Personality Styles from Patients' Speech [6.3042597209752715]
  We use natural language processing (NLP) and machine learning tools for classification. We test this on a dataset of recorded clinical diagnostic interviews (CDI) on a sample of 79 patients diagnosed with major depressive disorder (MDD) We find that automated classification with language-derived features (i.e., based on LIWC) significantly outperforms questionnaire-based classification models.
  arXiv  Detail & Related papers  (2023-11-07T15:56:19Z)
• PsyCoT: Psychological Questionnaire as Powerful Chain-of-Thought for Personality Detection [50.66968526809069]
  We propose a novel personality detection method, called PsyCoT, which mimics the way individuals complete psychological questionnaires in a multi-turn dialogue manner. Our experiments demonstrate that PsyCoT significantly improves the performance and robustness of GPT-3.5 in personality detection.
  arXiv  Detail & Related papers  (2023-10-31T08:23:33Z)
• Editing Personality for Large Language Models [73.59001811199823]
  This paper introduces an innovative task focused on editing the personality traits of Large Language Models (LLMs) We construct PersonalityEdit, a new benchmark dataset to address this task.
  arXiv  Detail & Related papers  (2023-10-03T16:02:36Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.