[2510.05174] Emergent Coordination in Multi-Agent Language Models

Computer Science > Multiagent Systems arXiv:2510.05174 (cs) [Submitted on 5 Oct 2025 (v1), last revised 27 Feb 2026 (this version, v2)] Title:Emergent Coordination in Multi-Agent Language Models Authors:Christoph Riedl View a PDF of the paper titled Emergent Coordination in Multi-Agent Language Models, by Christoph Riedl View PDF HTML (experimental) Abstract:When are multi-agent LLM systems merely a collection of individual agents versus an integrated collective with higher-order structure? We introduce an information-theoretic framework to test -- in a purely data-driven way -- whether multi-agent systems show signs of higher-order structure. This information decomposition lets us measure whether dynamical emergence is present in multi-agent LLM systems, localize it, and distinguish spurious temporal coupling from performance-relevant cross-agent synergy. We implement a practical criterion and an emergence capacity criterion operationalized as partial information decomposition of time-delayed mutual information (TDMI). We apply our framework to experiments using a simple guessing game without direct agent communication and minimal group-level feedback with three randomized interventions. Groups in the control condition exhibit strong temporal synergy but little coordinated alignment across agents. Assigning a persona to each agent introduces stable identity-linked differentiation. Combining personas with an instruction to ``think about what other agents might do'' shows i...