The Geometry of Dialogue: Graphing Language Models to Reveal Synergistic Teams for Multi-Agent Collaboration
This addresses the problem of automated team composition for multi-agent LLM collaboration, offering a practical solution for researchers and practitioners, though it is incremental as it builds on existing multi-agent and graph-based methods.
The paper tackles the challenge of forming optimal multi-agent teams of large language models (LLMs) by proposing an interaction-centric framework that constructs a 'language model graph' from pairwise conversations to identify synergistic clusters, resulting in teams that outperform random baselines and achieve comparable accuracy to manually-curated ones on downstream benchmarks.
While a multi-agent approach based on large language models (LLMs) represents a promising strategy to surpass the capabilities of single models, its success is critically dependent on synergistic team composition. However, forming optimal teams is a significant challenge, as the inherent opacity of most models obscures the internal characteristics necessary for effective collaboration. In this paper, we propose an interaction-centric framework for automatic team composition that does not require any prior knowledge including their internal architectures, training data, or task performances. Our method constructs a "language model graph" that maps relationships between models from the semantic coherence of pairwise conversations, and then applies community detection to identify synergistic model clusters. Our experiments with diverse LLMs demonstrate that the proposed method discovers functionally coherent groups that reflect their latent specializations. Priming conversations with specific topics identified synergistic teams which outperform random baselines on downstream benchmarks and achieve comparable accuracy to that of manually-curated teams based on known model specializations. Our findings provide a new basis for the automated design of collaborative multi-agent LLM teams.