AgentSocialBench demonstrates that privacy preservation is fundamentally harder in human-centered agentic social networks than in single-agent cases due to cross-domain coordination pressures and an abstraction paradox where privacy instructions increase discussion of sensitive information.
hub Canonical reference
S$^3$: Social-network Simulation System with Large Language Model-Empowered Agents
Canonical reference. 100% of citing Pith papers cite this work as background.
39
Pith papers citing it
316
external citations · Pith
Background
100% of classified citations
abstract
Social network simulation plays a crucial role in addressing various challenges within social science. It offers extensive applications such as state prediction, phenomena explanation, and policy-making support, among others. In this work, we harness the formidable human-like capabilities exhibited by large language models (LLMs) in sensing, reasoning, and behaving, and utilize these qualities to construct the S$^3$ system (short for $\textbf{S}$ocial network $\textbf{S}$imulation $\textbf{S}$ystem). Adhering to the widely employed agent-based simulation paradigm, we employ prompt engineering and prompt tuning techniques to ensure that the agent's behavior closely emulates that of a genuine human within the social network. Specifically, we simulate three pivotal aspects: emotion, attitude, and interaction behaviors. By endowing the agent in the system with the ability to perceive the informational environment and emulate human actions, we observe the emergence of population-level phenomena, including the propagation of information, attitudes, and emotions. We conduct an evaluation encompassing two levels of simulation, employing real-world social network data. Encouragingly, the results demonstrate promising accuracy. This work represents an initial step in the realm of social network simulation empowered by LLM-based agents. We anticipate that our endeavors will serve as a source of inspiration for the development of simulation systems within, but not limited to, social science.
hub tools
citation-role summary
background 9
citation-polarity summary
roles
background 9polarities
background 9representative citing papers
Introduces GenAI agent framework for auditing personalization algorithms via synthetic accounts with fixed personas, applied to X post-2024 election showing amplification of toxic and right-leaning content varying by ideology.
A hybrid simulator combining LLM decision-making with an explicit self-excitation model reproduces bursty temporal patterns in city-scale volunteering data, unlike pure LLM agents.
PERSUASIONTRACE introduces a Bayesian-network simulated target for multi-turn persuasion that matches human belief dynamics (81 vs 80) better than LLM baselines (64) and enables process-level evaluation.
ConsumerSimBench evaluates 13 LLMs on reconstructing crowd reactions from 1,553 Chinese social-media topics using 23,122 auditable yes-no criteria, finding maximum coverage of 47.8% by Gemini-3.1-Pro.
The paper unifies emerging graph-based world models under a new paradigm and proposes a taxonomy organized by spatial, physical, and logical relational inductive biases.
IntervenSim is an intervention-aware social network simulation that couples source interventions with crowd interactions in a feedback loop, improving MAPE by 41.6% and DTW by 66.9% over prior static frameworks on real-world events.
AgentMark watermarks agent planning behaviors with multi-bit identifiers via conditional sampling that preserves utility and works on black-box systems.
Graphia is an LLM post-training framework that uses real social graphs and GNN rewards to improve micro-level interaction prediction and macro-level network property replication in dynamic social simulations.
A systematic audit of LLM-based AI societies finds that 89.7% of 39 studies violate at least one of six PIMMUR validity principles, with reproductions showing that many claimed collective behaviors disappear when controls are tightened.
Autonomous LLM-agent networks form preferential-attachment structures with type-dependent centrality gaps, including capability-misaligned glass ceilings, predicted by a mean-field ODE and confirmed in 100-agent experiments.
A benchmark for LLM agents in partially observable joint decision-making reveals that deliberation challenges current models but can enable reflection and error correction.
MIDSim uses personalized LLM agents to jointly simulate social and algorithmic information diffusion streams and outperforms baselines on real diffusion events from three platforms.
Emergence World is a model-agnostic multi-agent simulation platform integrating live data, 120+ tools, persistent memory, and democratic governance, illustrated by a 15-day study showing divergent outcomes across five LLM models.
ES-MAS combines a new CURE dataset of 258 events and 14,000 news items with dual-stream integration and localized interaction modules to simulate opinion dynamics and claims better reproduction of historical U.S.-China attitude trends than prior simulators.
Language models show superior memory to humans on psych experiments but can be adjusted via prompting and compaction to forget more human-like, yielding better user simulators.
LLMs show mixed results on authorship verification, post generation, and attribute inference from Twitter data, with new frameworks and user studies establishing benchmarks for these analytics tasks.
SOCIA-EVO generates statistically consistent simulators by separating structural refinement from parameter calibration via bi-level optimization and falsifying strategies through execution feedback in a Bayesian-weighted playbook.
GraphMind equips LLM agents with graph awareness to construct human-like social networks, producing botnets that substantially degrade performance of both text-based and graph-based detectors.
LLM simulations of misinformation susceptibility overstate attitudinal associations and largely ignore personal network characteristics compared to human survey data.
This survey organizes generative recommendation into data, model, and task dimensions, identifying five advantages including world knowledge integration and creative generation while noting challenges in benchmarks and efficiency.
A two-phase data construction framework generates explanatory rationales from user feedback and applies uncertainty-based distillation to fine-tune lightweight LLMs as preference-aligned user simulators for recommender systems.
Introduces six-dimension trustworthiness definition and attention-based A-Trust score with a TMS to improve LLM-MAS robustness against malicious or unreliable messages.
AgentSociety is a large-scale LLM agent-based social simulator validated on polarization, UBI, disasters, and sustainability issues with alignment to real experiments.
citing papers explorer
-
AgentSocialBench: Evaluating Privacy Risks in Human-Centered Agentic Social Networks
AgentSocialBench demonstrates that privacy preservation is fundamentally harder in human-centered agentic social networks than in single-agent cases due to cross-domain coordination pressures and an abstraction paradox where privacy instructions increase discussion of sensitive information.
-
Using AI Agents to Automate Black-Box Audits of Personalization Algorithms at Scale
Introduces GenAI agent framework for auditing personalization algorithms via synthetic accounts with fixed personas, applied to X post-2024 election showing amplification of toxic and right-leaning content varying by ideology.
-
Toward Temporal Realism in City-Scale Crisis Response Simulation using LLM Agents
A hybrid simulator combining LLM decision-making with an explicit self-excitation model reproduces bursty temporal patterns in city-scale volunteering data, unlike pure LLM agents.
-
A Model of Multi-turn Human Persuadability Using Probabilistic Belief Tracing
PERSUASIONTRACE introduces a Bayesian-network simulated target for multi-turn persuasion that matches human belief dynamics (81 vs 80) better than LLM baselines (64) and enables process-level evaluation.
-
Can LLMs Think Like Consumers? Benchmarking Crowd-Level Reaction Reconstruction with ConsumerSimBench
ConsumerSimBench evaluates 13 LLMs on reconstructing crowd reactions from 1,553 Chinese social-media topics using 23,122 auditable yes-no criteria, finding maximum coverage of 47.8% by Gemini-3.1-Pro.
-
Graph World Models: Concepts, Taxonomy, and Future Directions
The paper unifies emerging graph-based world models under a new paradigm and proposes a taxonomy organized by spatial, physical, and logical relational inductive biases.
-
IntervenSim: Intervention-Aware Social Network Simulation for Opinion Dynamics
IntervenSim is an intervention-aware social network simulation that couples source interventions with crowd interactions in a feedback loop, improving MAPE by 41.6% and DTW by 66.9% over prior static frameworks on real-world events.
-
AgentMark: Utility-Preserving Behavioral Watermarking for Agents
AgentMark watermarks agent planning behaviors with multi-bit identifiers via conditional sampling that preserves utility and works on black-box systems.
-
GRAPHIA: Harnessing Social Graph Data to Enhance LLM-Based Social Simulation
Graphia is an LLM post-training framework that uses real social graphs and GNN rewards to improve micro-level interaction prediction and macro-level network property replication in dynamic social simulations.
-
The PIMMUR Principles: Ensuring Validity in Collective Behavior of LLM Societies
A systematic audit of LLM-based AI societies finds that 89.7% of 39 studies violate at least one of six PIMMUR validity principles, with reproductions showing that many claimed collective behaviors disappear when controls are tightened.
-
Emergence of Preferential Attachment and Glass-Ceiling Effects in Autonomous Networks of LLMs
Autonomous LLM-agent networks form preferential-attachment structures with type-dependent centrality gaps, including capability-misaligned glass ceilings, predicted by a mean-field ODE and confirmed in 100-agent experiments.
-
LLM Agents for Deliberative Collaboration: A Study on Joint Decision Making Under Partial Observability
A benchmark for LLM agents in partially observable joint decision-making reveals that deliberation challenges current models but can enable reflection and error correction.
-
MIDSim: Simulating Multi-Channel Information Diffusion in Social Media with LLM-Powered Multi-Agent System
MIDSim uses personalized LLM agents to jointly simulate social and algorithmic information diffusion streams and outperforms baselines on real diffusion events from three platforms.
-
Emergence World: A Platform for Evaluating Long-Horizon Multi-Agent Autonomy
Emergence World is a model-agnostic multi-agent simulation platform integrating live data, 120+ tools, persistent memory, and democratic governance, illustrated by a 15-day study showing divergent outcomes across five LLM models.
-
Modeling U.S. Attitudes Toward China via an Event-Steered Multi-Agent Simulator
ES-MAS combines a new CURE dataset of 258 events and 14,000 news items with dual-stream integration and localized interaction modules to simulate opinion dynamics and claims better reproduction of historical U.S.-China attitude trends than prior simulators.
-
Simulating Human Memory with Language Models
Language models show superior memory to humans on psych experiments but can be adjusted via prompting and compaction to forget more human-like, yielding better user simulators.
-
Assessing Capabilities of Large Language Models in Social Media Analytics: A Multi-task Quest
LLMs show mixed results on authorship verification, post generation, and attribute inference from Twitter data, with new frameworks and user studies establishing benchmarks for these analytics tasks.
-
SOCIA-EVO: Automated Simulator Construction via Dual-Anchored Bi-Level Optimization
SOCIA-EVO generates statistically consistent simulators by separating structural refinement from parameter calibration via bi-level optimization and falsifying strategies through execution feedback in a Bayesian-weighted playbook.
-
Beyond Individual Mimicry: Constructing Human-Like Social network with Graph-Augmented LLM Agents
GraphMind equips LLM agents with graph awareness to construct human-like social networks, producing botnets that substantially degrade performance of both text-based and graph-based detectors.
-
Overstating Attitudes, Ignoring Networks: LLM Biases in Simulating Misinformation Susceptibility
LLM simulations of misinformation susceptibility overstate attitudinal associations and largely ignore personal network characteristics compared to human survey data.
-
A Survey on Generative Recommendation: Data, Model, and Tasks
This survey organizes generative recommendation into data, model, and task dimensions, identifying five advantages including world knowledge integration and creative generation while noting challenges in benchmarks and efficiency.
-
Mirroring Users: Towards Building Preference-aligned User Simulator with User Feedback in Recommendation
A two-phase data construction framework generates explanatory rationales from user feedback and applies uncertainty-based distillation to fine-tune lightweight LLMs as preference-aligned user simulators for recommender systems.
-
To trust or not to trust: Attention-based Trust Management for LLM Multi-Agent Systems
Introduces six-dimension trustworthiness definition and attention-based A-Trust score with a TMS to improve LLM-MAS robustness against malicious or unreliable messages.
-
AgentSociety: Large-Scale Simulation of LLM-Driven Generative Agents Advances Understanding of Human Behaviors and Society
AgentSociety is a large-scale LLM agent-based social simulator validated on polarization, UBI, disasters, and sustainability issues with alignment to real experiments.
-
Cognitive Architectures for Language Agents
CoALA is a modular cognitive architecture for language agents that organizes memory components, action spaces for internal and external interaction, and a generalized decision-making loop to support more systematic development of capable agents.
-
A Survey on Large Language Model based Autonomous Agents
A survey of LLM-based autonomous agents that proposes a unified framework for their construction and reviews applications in social science, natural science, and engineering along with evaluation methods and future directions.
-
PerceptUI: LLM Agents as Human-Aligned Synthetic Users for UI/UX Evaluation
PerceptUI is a persona-conditioned LLM framework using contrastive reflection fine-tuning and prompt evolution to generate human-aligned UI/UX responses and rationales.
-
Mechanism Plausibility in Generative Agent-Based Modeling
Introduces the Mechanism Plausibility Scale, a four-level framework separating generative sufficiency from mechanistic plausibility in LLM-based agent-based models.
-
Topology-Aware LLM-Driven Social Simulation: A Unified Framework for Efficient and Realistic Agent Dynamics
TopoSim reduces LLM-based social simulation costs by 50-90% by grouping structurally similar agents for coordinated updates and using topology-derived PageRank to model asymmetric social influence.
-
Simulating Online Social Media Conversations on Controversial Topics Using AI Agents Calibrated on Real-World Data
LLM agents calibrated on Italian election data produce coherent posts and realistic network structure but show less tone and toxicity variation than real users, with opinion changes resembling traditional mathematical models.
-
From Human Memory to AI Memory: A Survey on Memory Mechanisms in the Era of LLMs
The paper surveys human memory categories, maps them to LLM memory, and proposes a new three-dimension (object, form, time) categorization into eight quadrants to organize existing work and highlight open problems.
-
A closer look at how large language models trust humans: patterns and biases
Across 43,200 simulations with five LLMs and five scenarios, model trust in humans aligns with human-like patterns driven by trustworthiness dimensions and is sometimes biased by age, gender, and religion.
-
AI as a Tool for Simulation-Based Experiments in Literary Studies
Proposes AI-driven simulations for literary-historical experiments and reports preliminary text-generation results claiming the first limited in-distribution outputs matching human novels.
-
Agentic World Modeling: Foundations, Capabilities, Laws, and Beyond
A survey proposing a three-level capability taxonomy (L1 Predictor, L2 Simulator, L3 Evolver) for world models across physical, digital, social, and scientific domains.
-
Network Effects and Agreement Drift in LLM Debates
LLM agents in controlled network debates show agreement drift toward specific opinion positions, requiring separation of structural effects from LLM biases before using them as human behavioral proxies.
-
Beyond Static Responses: Multi-Agent LLM Systems as a New Paradigm for Social Science Research
The paper maps LLM agent architectures onto a six-level continuum and argues that higher levels can enable simulation of emergent social phenomena while requiring attention to reproducibility and ethical issues.
-
Large Language Model based Multi-Agents: A Survey of Progress and Challenges
The paper surveys LLM-based multi-agent systems, covering simulated domains, agent profiling and communication, mechanisms for capacity growth, and common benchmarks.
-
Large Language Model Agent: A Survey on Methodology, Applications and Challenges
A survey that deconstructs LLM agent systems via a methodology-centered taxonomy linking design principles to emergent behaviors, applications, and challenges.
-
A Survey on the Memory Mechanism of Large Language Model based Agents
A systematic review of memory designs, evaluation methods, applications, limitations, and future directions for LLM-based agents.