Convergence of Outputs When Two Large Language Models Interact in a Multi-Agentic Setup

TL;DR

This paper investigates how two large language models interact in a multi-agent setup, revealing a tendency to converge into repetitive loops after initial coherent exchanges, regardless of their size or training.

Contribution

It demonstrates the convergence behavior of large language models in multi-agent interactions and introduces metrics to quantify this phenomenon.

Findings

Most conversations start coherently but later fall into repetition

Repetitive phrases emerge and dominate the dialogue

Models tend to produce similar outputs once repetition begins

Abstract

In this work, we report what happens when two large language models respond to each other for many turns without any outside input in a multi-agent setup. The setup begins with a short seed sentence. After that, each model reads the other's output and generates a response. This continues for a fixed number of steps. We used Mistral Nemo Base 2407 and Llama 2 13B hf. We observed that most conversations start coherently but later fall into repetition. In many runs, a short phrase appears and repeats across turns. Once repetition begins, both models tend to produce similar output rather than introducing a new direction in the conversation. This leads to a loop where the same or similar text is produced repeatedly. We describe this behavior as a form of convergence. It occurs even though the models are large, trained separately, and not given any prompt instructions. To study this behavior, we apply lexical and embedding-based metrics to measure how far the conversation drifts from the initial seed and how similar the outputs of the two models becomes as the conversation progresses.