Peer Influence across Heterogeneous AI Models

2026-10-02

AI Summary

This study addresses the poorly understood mechanisms of mutual persuasion among heterogeneous models in multi-agent systems, which render interaction outcomes difficult to predict. We propose a probabilistic decision shift metric that quantifies probability changes following single-round disagreement, enabling an empirical analysis of cross-model and cross-task interactions among seven open-source models on language understanding tasks. Our findings reveal that persuasiveness is determined by specific model pairings rather than by model scale or independence; notably, smaller models can overturn the judgments of larger ones. These results underscore the complexity of heterogeneous interaction dynamics, demonstrating that systemic behavior cannot be inferred from individual model attributes. Consequently, this work establishes a novel paradigm for evaluating multi-agent collaboration.

Abstract

When two AI agents disagree, who persuades whom? As multi-agent systems increasingly combine language models of different families and sizes, the answer can determine which judgments survive interaction. Measuring persuasion as the probabilistic shift in an agent's decision after a single exchange with a dissenting peer, we test seven open-weight models across three language understanding tasks. We find that persuasion is strong: when models disagree, receivers often abandon their initial judgment after seeing a peer's answer and explanation. Surprisingly, however, neither standalone certainty nor model scale reliably predicts persuasion dynamics. Models producing almost perfectly consistent decisions in isolation can be among the most susceptible to persuasion, and small models can match larger ones as persuaders and resist their influence just as effectively. Furthermore, we show that the size of the shift depends more on the susceptibility of the listener than on the persuasiveness of the speaker. Persuasion patterns are therefore specific to each model pairing, with heterogeneity amplifying persuasion in some combinations and suppressing it in others, allowing a dissenting agent running a small model to overturn the judgments of a much larger one. These findings show that the behavior of interacting models cannot be inferred from their individual properties but must be evaluated in the combinations in which they will operate.

Problem

• Peer Influence
• Heterogeneous AI Models
• Multi-agent Systems
• Persuasion Dynamics
• Language Models

Innovation

• Multi-agent systems
• Peer influence
• Heterogeneous models
• Persuasion dynamics
• Language models