Group size effects and collective misalignment in LLM multi-agent systems

arXiv:2510.22422 · cs.MA, cs.AI, cs.CY, physics.soc-ph · Submitted 2025-10-25 · Read on arXiv

cs.MA, cs.AI, cs.CY, physics.soc-ph

Submitted: 2025-10-25

Updated: 2025-10-25

Journal ref: Proc. Natl. Acad. Sci. U.S.A. 123 (34) e2531697123 (2026)

DOI: 10.1073/pnas.2531697123

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Mean-Field Dynamics — 11× in this paper · explained in 2 episodes
• Multi-Agent System — 6× in this paper · explained in 6 episodes
• Accuracy — 5× in this paper · explained in 1 episode
• Alignment — 5× in this paper · explained in 1 episode
• Robustness — 3× in this paper · explained in 5 episodes
• AI Agents — 2× in this paper · explained in 2 episodes
• Adam — 2× in this paper · explained in 2 episodes
• Cooperation — 2× in this paper · explained in 2 episodes
• Diffusion Process — 2× in this paper · explained in 2 episodes
• Dynamical regimes — 2× in this paper · explained in 2 episodes
• Hallucination — 2× in this paper · explained in 3 episodes
• Large Language Models (LLMs) — 2× in this paper · explained in 6 episodes
• Reward Function — 2× in this paper · explained in 2 episodes
• Transformer — 2× in this paper · explained in 2 episodes
• AI Alignment — 1× in this paper · explained in 2 episodes
• Actor — 1× in this paper · explained in 3 episodes
• Agentic AI — 1× in this paper · explained in 1 episode
• Elias — 1× in this paper · explained in 2 episodes
• Failure Modes — 1× in this paper · explained in 3 episodes
• In-Context Learning — 1× in this paper · explained in 4 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Steering — 1× in this paper · explained in 2 episodes

Sources

• Personal LLM Agents: Insights and Survey about the Capability, Efficiency and Security
• Large Language Model based Multi-Agents: A Survey of Progress and Challenges
• From Individual to Society: A Survey on Social Simulation Driven by Large Language Model-based Agents
• Debating with More Persuasive LLMs Leads to More Truthful Answers
• Many LLMs Are More Utilitarian Than One
• A Review of Cooperation in Multi-agent Learning
• Cultural evolution in populations of Large Language Models
• AI agents can coordinate beyond human scale
• LLM Agents Grounded in Self-Reports Enable General-Purpose Simulation of Individuals
• AgentSociety: Large-Scale Simulation of LLM-Driven Generative Agents Advances Understanding of Human Behaviors and Society
• OASIS: Open Agent Social Interaction Simulations with One Million Agents
• Modeling Earth-Scale Human-Like Societies with One Billion Agents
• The Coming Crisis of Multi-Agent Misalignment: AI Alignment Must Be a Dynamic and Social Process
• Hallucination is Inevitable: An Innate Limitation of Large Language Models
• A Survey on In-context Learning
• Large Language Models Cannot Self-Correct Reasoning Yet
• GTBench: Uncovering the Strategic Reasoning Limitations of LLMs via Game-Theoretic Evaluations
• Large Language Models Fail on Trivial Alterations to Theory-of-Mind Tasks
• Large Language Models Miss the Multi-Agent Mark

Related papers

• Highway Congestion Reduction through Reinforcement Learning Based Eulerian Headway Control
• You Only Align Once: Propagating Cooperative Behaviors in Multi-Agent Systems through Seed Agents
• Deny Without Disabling: Authorization-Paired Evaluation and Control for Multi-Agent Systems
• MA-SAPO: Multi-Agent Reasoning for Score-Aware Prompt Optimization
• PeroMAS: A Multi-agent System of Perovskite Material Discovery
• StitchCUDA: An Automated Multi-Agents End-to-End GPU Programing Framework with Rubric-based Agentic Reinforcement Learning