Strategic Communication and Language Bias in Multi-Agent LLM Coordination

TL;DR

This paper investigates how language framing and communication influence cooperation and biases in multi-agent LLM scenarios, revealing that communication can both promote coordination and reinforce language-driven biases.

Contribution

It demonstrates that communication impacts agent behavior differently across languages and game structures, highlighting the complex role of language in multi-agent coordination.

Findings

Communication significantly influences agent cooperation.

Language and personality affect communication impact.

Communication can reinforce biases in multi-agent interactions.

Abstract

Large Language Model (LLM)-based agents are increasingly deployed in multi-agent scenarios where coordination is crucial but not always assured. Research shows that the way strategic scenarios are framed linguistically can affect cooperation. This paper explores whether allowing agents to communicate amplifies these language-driven effects. Leveraging FAIRGAME, we simulate one-shot and repeated games across different languages and models, both with and without communication. Our experiments, conducted with two advanced LLMs-GPT-4o and Llama 4 Maverick-reveal that communication significantly influences agent behavior, though its impact varies by language, personality, and game structure. These findings underscore the dual role of communication in fostering coordination and reinforcing biases.