Mixture-of-Personas Language Models for Population Simulation
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866913780582055936 |
|---|---|
| author | Bui, Ngoc Nguyen, Hieu Trung Kumar, Shantanu Theodore, Julian Qiu, Weikang Nguyen, Viet Anh Ying, Rex |
| author_facet | Bui, Ngoc Nguyen, Hieu Trung Kumar, Shantanu Theodore, Julian Qiu, Weikang Nguyen, Viet Anh Ying, Rex |
| contents | Advances in Large Language Models (LLMs) paved the way for their emerging applications in various domains, such as human behavior simulations, where LLMs could augment human-generated data in social science research and machine learning model training. However, pretrained LLMs often fail to capture the behavioral diversity of target populations due to the inherent variability across individuals and groups. To address this, we propose \textit{Mixture of Personas} (MoP), a \textit{probabilistic} prompting method that aligns the LLM responses with the target population. MoP is a contextual mixture model, where each component is an LM agent characterized by a persona and an exemplar representing subpopulation behaviors. The persona and exemplar are randomly chosen according to the learned mixing weights to elicit diverse LLM responses during simulation. MoP is flexible, requires no model finetuning, and is transferable across base models. Experiments for synthetic data generation show that MoP outperforms competing methods in alignment and diversity metrics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_05019 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Mixture-of-Personas Language Models for Population Simulation Bui, Ngoc Nguyen, Hieu Trung Kumar, Shantanu Theodore, Julian Qiu, Weikang Nguyen, Viet Anh Ying, Rex Machine Learning Computation and Language Advances in Large Language Models (LLMs) paved the way for their emerging applications in various domains, such as human behavior simulations, where LLMs could augment human-generated data in social science research and machine learning model training. However, pretrained LLMs often fail to capture the behavioral diversity of target populations due to the inherent variability across individuals and groups. To address this, we propose \textit{Mixture of Personas} (MoP), a \textit{probabilistic} prompting method that aligns the LLM responses with the target population. MoP is a contextual mixture model, where each component is an LM agent characterized by a persona and an exemplar representing subpopulation behaviors. The persona and exemplar are randomly chosen according to the learned mixing weights to elicit diverse LLM responses during simulation. MoP is flexible, requires no model finetuning, and is transferable across base models. Experiments for synthetic data generation show that MoP outperforms competing methods in alignment and diversity metrics. |
| title | Mixture-of-Personas Language Models for Population Simulation |
| topic | Machine Learning Computation and Language |
| url | https://arxiv.org/abs/2504.05019 |