Conformity Dynamics in LLM Multi-Agent Systems: The Roles of Topology and Self-Social Weighting

2026-01-09

arXiv.org

AI Summary

This study investigates how network topology influences conformity behavior in large language model (LLM) multi-agent systems and its implications for judgment accuracy and robustness. By introducing a confidence-normalized pooling mechanism in a misinformation detection task, the work modulates the trade-off between individual self-reliance and social influence, and systematically compares centralized aggregation against distributed consensus paradigms. It reveals, for the first time, a conformity mechanism shaped by the interplay between network structure and the self–social weighting balance, identifying novel failure modes such as “confidently wrong” cascades. The findings show that centralized architectures enable efficient decision-making but are vulnerable to the central node’s capability and alignment biases among homogeneous models, whereas distributed structures offer greater robustness; however, high connectivity, while accelerating convergence, amplifies the risk of error propagation—providing critical theoretical guidance for designing LLM-based multi-agent systems.

Technology Category

• Multiagent Systems: Mechanism Design
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Large Language Models (LLMs) are increasingly instantiated as interacting agents in multi-agent systems (MAS), where collective decisions emerge through social interaction rather than independent reasoning. A fundamental yet underexplored mechanism in this process is conformity, the tendency of agents to align their judgments with prevailing group opinions. This paper presents a systematic study of how network topology shapes conformity dynamics in LLM-based MAS through a misinformation detection task. We introduce a confidence-normalized pooling rule that controls the trade-off between self-reliance and social influence, enabling comparisons between two canonical decision paradigms: Centralized Aggregation and Distributed Consensus. Experimental results demonstrate that network topology critically governs both the efficiency and robustness of collective judgments. Centralized structures enable immediate decisions but are sensitive to hub competence and exhibit same-model alignment biases. In contrast, distributed structures promote more robust consensus, while increased network connectivity speeds up convergence but also heightens the risk of wrong-but-sure cascades, in which agents converge on incorrect decisions with high confidence. These findings characterize the conformity dynamics in LLM-based MAS, clarifying how network topology and self-social weighting jointly shape the efficiency, robustness, and failure modes of collective decision-making.

Problem

• conformity
• multi-agent systems
• network topology
• collective decision-making
• LLM

Innovation

• conformity dynamics
• network topology
• self-social weighting
• LLM multi-agent systems
• confidence-normalized pooling