mlsanityData to 7 Oct 2026

Social simulation and multi-agent debate

1,066 papers9 this weekLow confidence in this grouping: the papers may belong to more than one topic, or the group may changeLow confidence in this grouping: the papers may belong to more than one topic, or the group may change · About

Keywords: agents, social, multi-agent, llm, llms

Key facts. Social simulation and multi-agent debate: 1,066 papers on arXiv since 2022 (data through 7 Oct 2026). 368 of its 1,066 mapped papers (35%) cite “Generative Agents: Interactive Simulacra of Human Behavior” (2023, arXiv 2304.03442). 131 of its 1,066 mapped papers (12%) cite “Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate” (2023, arXiv 2305.19118). 62 new papers 17 Sep–7 Oct.

What’s new

17 Sep – 7 Oct 2026

62 new papers, 17 Sep – 7 Oct; 9 this week.

  1. This week · 9
  2. Why LLM Agents Favor Their Group: Stakes, Observed Norms, and Reputation
  3. Back in Style: A Sociolinguistic Approach to Authoring and Measuring Persona Fidelity in User Simulation
  4. Decoupling Logic from Persona: Structural Immunity of Edge LLM Agents to Context Pollution also fits: General LLM agent frameworks and benchmarks
  5. Shared and structured inputs undermine collective random choice by reasoning AI agents
  6. From Retrieval to Customer Context: Evaluating Frontier-Model Systems for Voice-of-Customer Analysis also fits: Web, search, and research agents
  7. ToolRACER: A Robust Agentic Conversation Emulation Resource for Agent Training and Evaluation also fits: General LLM agent frameworks and benchmarks
  8. Wiki-Talkie: Multilingual Benchmarking of Persona-Based Agents on Real-World Discussions
  9. MASC: A Multi-Agent Self-Calibration Framework with Latent Construct Alignment for Consistent Client Role-Playing in Psychological Counseling
  10. On Open-Ended Information Seeking for Information Elicitation Agents
  11. Earlier · 53
  12. DREAM: Dynamic Resolution Assignment For Multimodal Multi-agent Debate also fits: Multimodal and embodied agents
All 62
  1. When Debate Helps: Proposal Supply and Verification-Aware Readout in Multi-Agent Reasoning
  2. Benchmarking Psychological Dynamics in Generative Agents
  3. Auditing Pairwise Equivalence Judgments: Self-Critique Effects and Diversity Measurement in Multi-Agent Hypothesis Generation also fits: Web, search, and research agents
  4. PB-GRPO: Learning Socially Adaptive LLM Agents from Persona-Driven Simulation with Preference-Batched GRPO
  5. Ask, Relax, or Act? Evaluating Actionable Indeterminacy in LLM Preference Reasoning also fits: General LLM agent frameworks and benchmarks
  6. PsyEvo: A Personalized Counseling Agent That Self-Evolves at Test Time also fits: Medical and scientific domain agents
  7. Silent Dissent: LLM Agents That Yield to the Majority Still Represent Their Original Premise
  8. Coherence-Driven Belief Formation and Population Dynamics of Contagion in LLM Agents
  9. Before Agents Decide: Epistemic Action in LLM-Based Systems also fits: General LLM agent frameworks and benchmarks
  10. JuryFlow: Disagreement-Guided Human-in-the-Loop Multi-Agent Evaluation also fits: Web, search, and research agents
  11. MADBench: Benchmarking the Security of Multi-Agent Debate
  12. When Order Matters: First-Speaker Bias and Mitigation through Personality in Sequential Multi-Agent Debate
  13. Persona and Persuasive Framing in AI Voice Agents: A 2×2 Field Experiment with Children
  14. From Solo to Social Learning: Characterizing Recursive Social Improvement in LLMs also fits: Reinforcement learning for LLM agents
  15. Multi-agent discussion gains less when dissent is withheld
  16. AI Agents are Vulnerable to Radicalization
  17. When Upstream Messages Override Correct Answers: A Controlled Study of Multi-Agent LLM Collaboration
  18. Engineering Simplicity: Simple Mechanism Interfaces Steer LLM Agents also fits: General LLM agent frameworks and benchmarks
  19. PADMÉ: Preference Alignment Data Synthesis for Meta-Evaluation of LM Agent Evaluators
  20. Prompted Identity Degrades Cooperation in Multi-Agent LLM Systems
  21. Measuring Collapse and Correction in Homogeneous-Panel LLM Debate
  22. From Preference to Reciprocity: Decentralized Matching with Empirically Grounded LLM-agent Based Modeling
  23. Social Circuits behind Multi-agent Echo Chambers
  24. Cheap, open agents make LLM pollution harder to mitigate also fits: Agent threats, guardrails and governance
  25. Designing Reliable LLM-as-a-Judge Measurement Systems for Multi-Turn Business Agents also fits: Software engineering coding agents
  26. LLMs Trust Their Own: Identity-Dependent Conformity in Multi-Agent Systems
  27. Beyond Symmetric Agents: Cognitive Diversity and Multi-Agent Debate in Small Language Models
  28. Trust and Task Completion in the World of Consumer AI Agents
  29. AgentHabit: Characterizing Distinct Behaviors of Agents on Everyday Tasks also fits: General LLM agent frameworks and benchmarks
  30. On the Behavioral Traits of LLM Agents
  31. On Evaluating and Improving Conversational Agents in Production
  32. Towards Mitigating Fabricated Consensus: The Active Provenance Gate for Multi-Agent Debate Synthesis also fits: Web, search, and research agents
  33. Thinking Less to Simulate Better: Intuitive Prompting Improves LLM Agents Simulating Individual Social Media Reactions, Including Unfamiliar Content
  34. Working with Agentic `Teammates': When a New Organizational Actor Collides with the Human Ecosystem of Work also fits: General LLM agent frameworks and benchmarks
  35. Persuaded, Not Informed: Incentive-Misaligned Witnesses Defeat In-Context Grounding also fits: Financial trading and forecasting agents
  36. Agent-based Modeling: Equilibrium, Echo Chambers, and Efficiency in Hybrid Coevolutionary Opinion Games
  37. CAVEAT: Towards Robust Computer-Use Agents in Incentive-Misaligned Environments also fits: Agent threats, guardrails and governance
  38. Tipping Points in LLM-Based Multi-Agent Systems: Stance on Climate Change Action
  39. When LLM Agents Fail to Read the Room: ReAdapt for Relational Social Reasoning
  40. Emergent Collusion in Long-Horizon LLM Agent Interaction
  41. SocioVerse2: A Longitudinal Dynamic Social Simulation Framework under a Human-AI Co-evolutionary Paradigm
  42. Small-world Networks of Agents Brainstorm AI Risks to Support Ideation
  43. Mind or Message? Auditing Theory of Mind in Multi-Agent Social Simulation
  44. Local Predictability and Collective Fidelity in LLM-Agent Societies
  45. Be Careful Who You Trust: Coordination Dynamics under Corrupted Communication in LLM Multi-Agent Games
  46. From Certain Doom to Survival: Agent-Driven Self-Governance in LLM Agent Societies
  47. Value-Sensitive Delegation in Everyday AI Agent Use: Evidence from OpenClaw also fits: Agent threats, guardrails and governance
  48. Bayesian Belief Layer for Controllable Opinion Dynamics in LLM Agents
  49. Do Personality-Tuned LLMs Make Better Social Agents?
  50. A Dual-Process Perspective on Nudge Susceptibility in LLM-Based GUI Agents
  51. Faithful Where It Can Be Checked: Auditing a Reflection Agent Against Its System Prompt in a Randomized Trial also fits: Educational agents and math problem solving
  52. When Hiring Becomes Agent-Mediated: Evaluating Access and Recurrence in Two-Agent R'esum'e Screening

Foundations

Most cited by papers here

  1. Generative Agents: Interactive Simulacra of Human Behavior2023cited by 368 here
  2. Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate2023cited by 131 here
  3. AgentSociety: Large-Scale Simulation of LLM-Driven Generative Agents Advances Understanding of Human Behaviors and Society2025cited by 62 here
  4. Character-LLM: A Trainable Agent for Role-Playing2023cited by 60 here
  5. Exploring Collaboration Mechanisms for LLM Agents: A Social Psychology View2023cited by 53 here

Lineage

The papers this direction builds on, oldest first.

Each paper on the line cites the one before.

  1. 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 368 here
  2. 2023MetaGPT: Meta Programming for A Multi-Agent Collaborative Frameworkcited by 99 here
  3. 2023AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversationcited by 107 here
  4. 2023A Survey on Large Language Model based Autonomous Agentscited by 95 here
  5. 2023The Rise and Potential of Large Language Model Based Agents: A Surveycited by 59 here
  6. 2023Large Language Models Empowered Agent-based Modeling and Simulation: A Survey and Perspectivescited by 58 here
  7. 2024Can Large Language Model Agents Simulate Human Trust Behavior?cited by 45 here
17 more it builds on, not in this chain
  1. 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 45 here
  2. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 77 here
  3. 2023CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Societycited by 94 here
  4. 2023Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debatecited by 131 here
  5. 2023ChatDev: Communicative Agents for Software Developmentcited by 44 here
  6. 2023ChatEval: Towards Better LLM-based Evaluators through Multi-Agent Debatecited by 108 here
  7. 2023AgentVerse: Facilitating Multi-Agent Collaboration and Exploring Emergent Behaviorscited by 57 here
  8. 2023Exploring Collaboration Mechanisms for LLM Agents: A Social Psychology Viewcited by 53 here
  9. 2023Character-LLM: A Trainable Agent for Role-Playingcited by 60 here
  10. 2023EconAgent: Large Language Model-Empowered Agents for Simulating Macroeconomic Activitiescited by 51 here
  11. 2023SOTOPIA: Interactive Evaluation for Social Intelligence in Language Agentscited by 54 here
  12. 2023Simulating Opinion Dynamics with Networks of LLM-based Agentscited by 51 here
  13. 2023Should we be going MAD? A Look at Multi-Agent Debate Strategies for LLMscited by 37 here
  14. 2024Large Language Model based Multi-Agents: A Survey of Progress and Challengescited by 47 here
  15. 2024Cooperate or Collapse: Emergence of Sustainable Cooperation in a Society of LLM Agentscited by 42 here
  16. 2024LLM Agents Grounded in Self-Reports Enable General-Purpose Simulation of Individualscited by 39 here
  17. 2025AgentSociety: Large-Scale Simulation of LLM-Driven Generative Agents Advances Understanding of Human Behaviors and Societycited by 62 here

The 24 most cited of 1,521.

Browse

30 of 1,066, ranked 6–35

  1. 6EconAgent: Large Language Model-Empowered Agents for Simulating Macroeconomic Activities202351
  2. 7Simulating Opinion Dynamics with Networks of LLM-based Agents202351
  3. 8Can Large Language Model Agents Simulate Human Trust Behavior?202445
  4. 9Cooperate or Collapse: Emergence of Sustainable Cooperation in a Society of LLM Agents202442
  5. 10LLM Agents Grounded in Self-Reports Enable General-Purpose Simulation of Individuals202439
  6. 11Should we be going MAD? A Look at Multi-Agent Debate Strategies for LLMs202337
  7. 12CharacterEval: A Chinese Benchmark for Role-Playing Conversational Agent Evaluation202435
  8. 13From Individual to Society: A Survey on Social Simulation Driven by Large Language Model-based Agents202435
  9. 14From Persona to Personalization: A Survey on Role-Playing Language Agents202431
  10. 15War and Peace (WarAgent): Large Language Model-based Multi-Agent Simulation of World Wars202329
All 30
  1. 16GroupDebate: Enhancing the Efficiency of Multi-Agent Debate Using Group Discussion202428
  2. 17User Behavior Simulation with Large Language Model based Agents202328
  3. 18Generative agent-based modeling with actions grounded in physical, social, or digital space using Concordia202327
  4. 19Debate or Vote: Which Yields Better Decisions in Multi-Agent Large Language Models?202526
  5. 20Unveiling the Truth and Facilitating Change: Towards Agent-based Large-scale Social Movement Simulation202424
  6. 21GenSim: A General Social Simulation Platform with Large Language Model based Agents202424
  7. 22Talk Isn't Always Cheap: Understanding Failure Modes in Multi-Agent Debate202522
  8. 23How Well Can LLMs Negotiate? NegotiationArena Platform and Analysis202421
  9. 24Voting or Consensus? Decision-Making in Multi-Agent Debate202519
  10. 25Epidemic Modeling with Generative Agents202319
  11. 26Systematic Biases in LLM Simulations of Debates202418
  12. 27Shall We Team Up: Exploring Spontaneous Cooperation of Competing LLM Agents202417
  13. 28CompeteAI: Understanding the Competition Dynamics in Large Language Model-based Agents202317
  14. 29MetaAgents: Large Language Model Based Agents for Decision-Making on Teaming202316
  15. 30Do LLM Agents Exhibit Social Behavior?202316
  16. 31AgentsCourt: Building Judicial Decision-Making Agents with Court Debate Simulation and Legal Knowledge Augmentation202415
  17. 32PersonaGym: Evaluating Persona Agents and LLMs202415
  18. 33Humanoid Agents: Platform for Simulating Human-like Generative Agents202315
  19. 34Stop Overvaluing Multi-Agent Debate -- We Must Rethink Evaluation and Embrace Model Heterogeneity202514
  20. 35Emergence of Social Norms in Generative Agent Societies: Principles and Architecture202414