# PolicySim: An LLM-Based Agent Social Simulation Sandbox for Proactive Policy Optimization

arXiv: https://arxiv.org/abs/2603.19649 · HTML: https://mlsanity.com/paper/2603.19649/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Placed in Social simulation and multi-agent debate, one of 16 directions in LLM agents and tool use.
- Builds on 34 of its 75 references that are in the map.
- First posted on arXiv 20 Mar 2026.

## Placed in

- LLM agents and tool use / [Social simulation and multi-agent debate](https://mlsanity.com/agents/social-simulation-and-multi-agent-debate/)

## Builds on (most cited first, 30 of 34)

- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation](https://mlsanity.com/paper/2002.02126/) (2020)
- [ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools](https://mlsanity.com/paper/2406.12793/) (2024)
- [AgentVerse: Facilitating Multi-Agent Collaboration and Exploring Emergent Behaviors](https://mlsanity.com/paper/2308.10848/) (2023)
- [Social Simulacra: Creating Populated Prototypes for Social Computing Systems](https://mlsanity.com/paper/2208.04024/) (2022)
- [Towards Deeper Graph Neural Networks](https://mlsanity.com/paper/2007.09296/) (2020)
- [S3: Social-network Simulation System with Large Language Model-Empowered Agents](https://mlsanity.com/paper/2307.14984/) (2023)
- [A New Generation of Perspective API: Efficient Multilingual Character-level Transformers](https://mlsanity.com/paper/2202.11176/) (2022)
- [EconAgent: Large Language Model-Empowered Agents for Simulating Macroeconomic Activities](https://mlsanity.com/paper/2310.10436/) (2023)
- [Simulating Opinion Dynamics with Networks of LLM-based Agents](https://mlsanity.com/paper/2311.09618/) (2023)
- [OASIS: Open Agent Social Interaction Simulations with One Million Agents](https://mlsanity.com/paper/2411.11581/) (2024)
- [Offline A/B testing for Recommender Systems](https://mlsanity.com/paper/1801.07030/) (2018)
- [On Generative Agents in Recommendation](https://mlsanity.com/paper/2310.10108/) (2023)
- [Unveiling the Truth and Facilitating Change: Towards Agent-based Large-scale Social Movement Simulation](https://mlsanity.com/paper/2402.16333/) (2024)
- [Simulating Social Media Using Large Language Models to Evaluate Alternative News Feed Algorithms](https://mlsanity.com/paper/2310.05984/) (2023)
- [Stance Detection with Collaborative Role-Infused LLM-Based Agents](https://mlsanity.com/paper/2310.10467/) (2023)
- [TwiBot-20: A Comprehensive Twitter Bot Detection Benchmark](https://mlsanity.com/paper/2106.13088/) (2021)
- [EE-Net: Exploitation-Exploration Neural Networks in Contextual Bandits](https://mlsanity.com/paper/2110.03177/) (2021)
- [The Wisdom of Partisan Crowds: Comparing Collective Intelligence in Humans and LLM-based Agents](https://mlsanity.com/paper/2311.09665/) (2023)
- [DGCN: Diversified Recommendation with Graph Convolutional Networks](https://mlsanity.com/paper/2108.06952/) (2021)
- [Digital Nudging with Recommender Systems: Survey and Future Directions](https://mlsanity.com/paper/2011.03413/) (2020)
- [LMAgent: A Large-scale Multimodal Agents Society for Multi-user Simulation](https://mlsanity.com/paper/2412.09237/) (2024)
- [A Graph-based Approach for Mitigating Multi-sided Exposure Bias in Recommender Systems](https://mlsanity.com/paper/2107.03415/) (2021)
- [Better with Less: A Data-Active Perspective on Pre-Training Graph Neural Networks](https://mlsanity.com/paper/2311.01038/) (2023)
- [MF-LLM: Simulating Population Decision Dynamics via a Mean-Field Large Language Model Framework](https://mlsanity.com/paper/2504.21582/) (2025)
- [A Hierarchical Location Prediction Neural Network for Twitter User Geolocation](https://mlsanity.com/paper/1910.12941/) (2019)
- [Graph Neural Network Bandits](https://mlsanity.com/paper/2207.06456/) (2022)
- [SOTOPIA-S4: a user-friendly system for flexible, customizable, and large-scale social simulation](https://mlsanity.com/paper/2504.16122/) (2025)

## Built on by (most cited first, 10 of 10)

- [Agentic World Modeling: Foundations, Capabilities, Laws, and Beyond](https://mlsanity.com/paper/2604.22748/) (2026)
- [Stop Drawing Scientific Claims from LLM Social Simulations Without Robustness Audits](https://mlsanity.com/paper/2605.18890/) (2026)
- [RAT: RunAnyThing via Fully Automated Environment Configuration](https://mlsanity.com/paper/2604.23190/) (2026)
- [AgoraSim: A Hybrid Agent-Based Modeling Framework](https://mlsanity.com/paper/2607.05999/) (2026)
- [Cyberbullying Governance on Social Media: A Unified Framework from Content Identification to Intervention](https://mlsanity.com/paper/2605.27584/) (2026)
- [PTCG-Bench: Can LLM Agents Master Pok'emon Trading Card Game?](https://mlsanity.com/paper/2605.29653/) (2026)
- [RedditPersona: A Modular Framework for Community-Conditioned LLM Adaptation from Reddit](https://mlsanity.com/paper/2606.06027/) (2026)
- [CoRenew: A large language model agent-based policy simulation platform for multifamily residential redevelopment](https://mlsanity.com/paper/2607.25447/) (2026)
- [LiveSim: Simulating Environment-Shaped Users in Multi-Agent Live-Stream Ecosystems](https://mlsanity.com/paper/2608.26849/) (2026)
- [LLMs for Social Network Modeling: From Network Generation to Dynamic Processes](https://mlsanity.com/paper/2609.08049/) (2026)

## Abstract (arXiv)

Social platforms serve as central hubs for information exchange, where user behaviors and platform interventions jointly shape opinions. However, intervention policies like recommendation and content filtering, can unintentionally amplify echo chambers and polarization, posing significant societal risks. Proactively evaluating the impact of such policies is therefore crucial. Existing approaches primarily rely on reactive online A/B testing, where risks are identified only after deployment, making risk identification delayed and costly. LLM-based social simulations offer a promising pre-deployment alternative, but current methods fall short in realistically modeling platform interventions and incorporating feedback from the platform. Bridging these gaps is essential for building actionable frameworks to assess and optimize platform policies. To this end, we propose PolicySim, an LLM-based social simulation sandbox for the proactive assessment and optimization of intervention policies. PolicySim models the bidirectional dynamics between user behavior and platform interventions through two key components: (1) a user agent module refined via supervised fine-tuning (SFT) and direct preference optimization (DPO) to achieve platform-specific behavioral realism; and (2) an adaptive intervention module that employs a contextual bandit with message passing to capture dynamic network structures. Experiments show that PolicySim can accurately simulate platform ecosystems at both micro and macro levels and support effective intervention policy.