# Mixture-of-Personas Language Models for Population Simulation

arXiv: https://arxiv.org/abs/2504.05019 · HTML: https://mlsanity.com/paper/2504.05019/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 68 references that are in the map.
- First posted on arXiv 7 Apr 2025.

## Builds on (most cited first, 19 of 19)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [High-Dimensional Continuous Control Using Generalized Advantage Estimation](https://mlsanity.com/paper/1506.02438/) (2015)
- [Multi-Agent Deep Reinforcement Learning for Large-scale Traffic Signal Control](https://mlsanity.com/paper/1903.04527/) (2019)
- [Deep Reinforcement Learning for Intelligent Transportation Systems: A Survey](https://mlsanity.com/paper/2005.00935/) (2020)
- [Learning Phase Competition for Traffic Signal Control](https://mlsanity.com/paper/1905.04722/) (2019)
- [Traffic Light Control Using Deep Policy-Gradient and Value-Function Based Reinforcement Learning](https://mlsanity.com/paper/1704.08883/) (2017)
- [Expression might be enough: representing pressure and demand for reinforcement learning based traffic signal control](https://mlsanity.com/paper/2112.10107/) (2021)
- [Using a Deep Reinforcement Learning Agent for Traffic Signal Control](https://mlsanity.com/paper/1611.01142/) (2016)
- [Can ChatGPT Enable ITS? The Case of Mixed Traffic Control via Reinforcement Learning](https://mlsanity.com/paper/2306.08094/) (2023)
- [Learning to Control and Coordinate Mixed Traffic Through Robot Vehicles at Complex and Unsignalized Intersections](https://mlsanity.com/paper/2301.05294/) (2023)
- [Large-scale Mixed Traffic Control Using Dynamic Vehicle Routing and Privacy-Preserving Crowdsourcing](https://mlsanity.com/paper/2311.11347/) (2023)
- [Mixed Traffic Control and Coordination from Pixels](https://mlsanity.com/paper/2302.09167/) (2023)
- [EnduRL: Enhancing Safety, Stability, and Efficiency of Mixed Traffic Under Real-World Perturbations Via Reinforcement Learning](https://mlsanity.com/paper/2311.12261/) (2023)
- [Black-box Adversarial Attacks on Network-wide Multi-step Traffic State Prediction Models](https://mlsanity.com/paper/2110.08712/) (2021)
- [CARL: Congestion-Aware Reinforcement Learning for Imitation-based Perturbations in Mixed Traffic Control](https://mlsanity.com/paper/2404.00796/) (2024)
- [Heterogeneous Mixed Traffic Control and Coordination](https://mlsanity.com/paper/2409.12330/) (2024)
- [Time-to-Green predictions for fully-actuated signal control systems with supervised learning](https://mlsanity.com/paper/2208.11344/) (2022)
- [Deep Reinforcement Learning for Traffic Light Control in Intelligent Transportation Systems](https://mlsanity.com/paper/2302.03669/) (2023)
- [A pedestrian hopping model and traffic light scheduling for pedestrian-vehicle mixed-flow networks](https://mlsanity.com/paper/1705.05251/) (2017)

## Built on by (most cited first, 20 of 20)

- [SMART: Simulated Students Aligned with Item Response Theory for Question Difficulty Prediction](https://mlsanity.com/paper/2507.05129/) (2025)
- [Valid Survey Simulations with Limited Human Data: The Roles of Prompting, Fine-Tuning, and Rectification](https://mlsanity.com/paper/2510.11408/) (2025)
- [Persona-Based Synthetic Data Generation Using Multi-Stage Conditioning with Large Language Models for Emotion Recognition](https://mlsanity.com/paper/2507.13380/) (2025)
- [PersonaFuse: A Personality Activation-Driven Framework for Enhancing Human-LLM Interactions](https://mlsanity.com/paper/2509.07370/) (2025)
- [CrowdLLM: Building LLM-Based Digital Populations Augmented with Generative Models](https://mlsanity.com/paper/2512.07890/) (2025)
- [Structured Pruning for Diverse Best-of-N Reasoning Optimization](https://mlsanity.com/paper/2506.03978/) (2025)
- [Do Persona-Infused LLMs Affect Performance in a Strategic Reasoning Game?](https://mlsanity.com/paper/2512.06867/) (2025)
- [Reasoning Planning for Language Models](https://mlsanity.com/paper/2511.00521/) (2025)
- [SYN-DIGITS: A Synthetic Control Framework for Calibrated Digital Twin Simulation](https://mlsanity.com/paper/2604.07513/) (2026)
- [IROTE: Human-like Traits Elicitation of Large Language Model via In-Context Self-Reflective Optimization](https://mlsanity.com/paper/2508.08719/) (2025)
- [From Model Training to Model Raising](https://mlsanity.com/paper/2511.09287/) (2025)
- [HAG: Hierarchical Demographic Tree-based Agent Generation for Topic-Adaptive Simulation](https://mlsanity.com/paper/2601.05656/) (2026)
- [RADIUS: Ranking, Distribution, and Significance - A Comprehensive Alignment Suite for Survey Simulation](https://mlsanity.com/paper/2603.19002/) (2026)
- [KnowledgeGain: Evaluating and Optimizing Science News Generation for Reader Learning](https://mlsanity.com/paper/2605.31099/) (2026)
- [Distribution-First Population Simulation: Collapse, Calibration, and Recall in Non-WEIRD LLM Persona Modeling](https://mlsanity.com/paper/2607.18310/) (2026)
- [Data-Driven Persona-Conditioned Agents for A/B Test Simulation](https://mlsanity.com/paper/2609.01038/) (2026)
- [LLM Active Alignment: A Nash Equilibrium Perspective](https://mlsanity.com/paper/2602.06836/) (2026)
- [Adaptive Querying with AI Persona Priors](https://mlsanity.com/paper/2605.00696/) (2026)
- [LLM-Powered Virtual Population for Demand Simulation and Pricing](https://mlsanity.com/paper/2606.16183/) (2026)
- [AppLooper: An Agentic Application Engineering Loop for Accountable Release with Virtual-User Feedback](https://mlsanity.com/paper/2608.14093/) (2026)

## Abstract (arXiv)

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 textitMixture of Personas (MoP), a textitprobabilistic 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.