mlsanityData to 7 Oct 2026

Multi-Agent Systems are Mixtures of Experts: Who Becomes an Influencer?

arXiv 2605.25929 ↗First posted 25 May 2026cs.MA cs.LGcited by 0

On the map: LLM agents and tool use → Social simulation and multi-agent debate

Abstract (arXiv)

The effectiveness of multi-agent LLM deliberation depends not only on the agents' individual predictions, but also on how they communicate and collaborate. We study this mechanism through the lens of Friedkin-Johnsen (FJ) opinion dynamics, a tractable model for analyzing stubbornness, influence, and opinion change in multi-agent systems that captures empirically observed deliberation patterns. We show that the FJ parameters are input-dependent, turning multi-agent deliberation into a mixture of experts. This perspective implies that multi-agent systems can outperform single agents and static ensembles when routing reflects agent competence. Since competence is latent in practice, we analyze how influence is established through observable proxies: agents' self-assessed confidence, their perceived confidence, and initial alignment with other agents' views.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Social simulation and multi-agent debate, one of 16 directions in LLM agents and tool use. Builds on 39 of its 62 references that are in the map. First posted on arXiv 25 May 2026.

Builds on

39 of 62 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025Stop Overvaluing Multi-Agent Debate -- We Must Rethink Evaluation and Embrace Model Heterogeneity12 Feb
  2. 2023Should we be going MAD? A Look at Multi-Agent Debate Strategies for LLMs29 Nov

Other references · 37

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  2. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,601
  3. 2019A Comprehensive Survey on Graph Neural Networkscited by 2,906
  4. 2023AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversationcited by 1,906
  5. 2018CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledgecited by 1,793
  6. 2023MetaGPT: Meta Programming for A Multi-Agent Collaborative Frameworkcited by 1,761
  7. 2024MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmarkcited by 1,758
  8. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,739
  9. 2023CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Societycited by 1,348
  10. 2023ChatDev: Communicative Agents for Software Developmentcited by 893
  11. 2021How Attentive are Graph Attention Networks?cited by 818
  12. 2021BBQ: A Hand-Built Bias Benchmark for Question Answeringcited by 683
25 more
  1. 2025Why Do Multi-Agent LLM Systems Fail?cited by 559
  2. 2023Better Zero-Shot Reasoning with Role-Play Promptingcited by 220
  3. 2024A Survey on LLM-based Multi-Agent System: Recent Advances and New Frontiers in Applicationcited by 54
  4. 2019Explainability in Human-Agent Systemscited by 34
  5. 2024Decoding Echo Chambers: LLM-Powered Simulations Revealing Polarization in Social Networkscited by 31
  6. 2026Single-Agent LLMs Outperform Multi-Agent Systems on Multi-Hop Reasoning Under Equal Thinking Token Budgetscited by 28
  7. 2024Spectral Graph Pruning Against Over-Squashing and Over-Smoothingcited by 23
  8. 2017A Tutorial on Modeling and Analysis of Dynamic Social Networks. Part Icited by 15
  9. 2025GNNs Getting ComFy: Community and Feature Similarity Guided Rewiringcited by 13
  10. 2015Novel Multidimensional Models of Opinion Dynamics in Social Networkscited by 10
  11. 2023Are GATs Out of Balance?cited by 9
  12. 2026Expert Personas Improve LLM Alignment but Damage Accuracy: Bootstrapping Intent-Based Persona Routing with PRISMcited by 9
  13. 2024Preventing Representational Rank Collapse in MPNNs by Splitting the Computational Graphcited by 6
  14. 2025When Shift Happens - Confounding Is to Blamecited by 5
  15. 2023On the Two Sides of Redundancy in Graph Neural Networkscited by 4
  16. 2024GATE: How to Keep Out Intrusive Neighborscited by 4
  17. 2026Opinion dynamics and mutual influence with LLM agents through dialog simulationcited by 4
  18. 2026Don't Trust Stubborn Neighbors: A Security Framework for Agentic Networkscited by 3
  19. 2016Opinion dynamics in social networks with stubborn agents: an issue-based perspectivecited by 2
  20. 2026Fixed Aggregation Features Can Rival GNNscited by 2
  21. 2026Social Dynamics as Critical Vulnerabilities that Undermine Objective Decision-Making in LLM Collectivescited by 2
  22. 2017A framework for cascade size calculations on random networkscited by 1
  23. 2021Scaling up Continuous-Time Markov Chains Helps Resolve Underspecificationcited by 1
  24. 2025Towards Ethical Multi-Agent Systems of Large Language Models: A Mechanistic Interpretability Perspectivecited by 1
  25. 2026Robustness of Mixtures of Experts to Feature Noisecited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Multi-Agent Consensus as a Cognitive Bias Trigger in Human-AI Interaction
  2. 2026Hidden Anchors in Multi-Agent LLM Deliberation
  3. 2026Belief Cascades Drive Persuasion in LLM Agent Networks
  4. 2025An Empirical Study of Group Conformity in Multi-Agent Systems
  5. 2026The Deliberative Illusion: Diagnosing Factual Attrition and Stance Homogenization in Multi-Agent LLM Deliberation
  6. 2026Physics of Agents: Statistical Mechanics Predicts Collective Behavior of AI Agents
  7. 2026Conformity Dynamics in LLM Multi-Agent Systems: The Roles of Topology and Self-Social Weighting
  8. 2026LLM Agents Make Collective Belief Dynamics Programmable: Challenges and Research Directions