When Numbers Start Talking: Implicit Numerical Coordination Among LLM-Based Agents

TL;DR

This paper investigates how LLM-based agents coordinate implicitly through covert signals in multi-agent environments, analyzing the emergence and impact of such signals across various game-theoretic settings.

Contribution

It provides a game-theoretic analysis of covert communication among LLM agents, highlighting conditions for its emergence and influence on strategic outcomes.

Findings

Covert signals can emerge in multi-agent interactions without explicit communication.

Different communication regimes significantly affect coordination outcomes.

Heterogeneous agent personalities influence the likelihood and nature of covert signaling.

Abstract

LLMs-based agents increasingly operate in multi-agent environments where strategic interaction and coordination are required. While existing work has largely focused on individual agents or on interacting agents sharing explicit communication, less is known about how interacting agents coordinate implicitly. In particular, agents may engage in covert communication, relying on indirect or non-linguistic signals embedded in their actions rather than on explicit messages. This paper presents a game-theoretic study of covert communication in LLM-driven multi-agent systems. We analyse interactions across four canonical game-theoretic settings under different communication regimes, including explicit, restricted, and absent communication. Considering heterogeneous agent personalities and both one-shot and repeated games, we characterise when covert signals emerge and how they shape coordination and strategic outcomes.