Guardado en:
Detalles Bibliográficos
Autores principales: Furuya, Kotaro, Kitagawa, Yuichi
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2510.26352
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866918396733423616
author Furuya, Kotaro
Kitagawa, Yuichi
author_facet Furuya, Kotaro
Kitagawa, Yuichi
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2510_26352
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle The Geometry of Dialogue: Graphing Language Models to Reveal Synergistic Teams for Multi-Agent Collaboration
Furuya, Kotaro
Kitagawa, Yuichi
Computation and Language
Artificial Intelligence
Multiagent Systems
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.
title The Geometry of Dialogue: Graphing Language Models to Reveal Synergistic Teams for Multi-Agent Collaboration
topic Computation and Language
Artificial Intelligence
Multiagent Systems
url https://arxiv.org/abs/2510.26352