What question did this study set out to answer?

To evaluate whether structured multi-agent debate frameworks and model diversity improve reasoning reliability and decision quality in high-stakes satellite communication tasks.

August 15, 2026Journal of Intelligent Information Systems0 citationsOpen Access

Multi-Agent debate system based on large language models: structured deliberation and validation in satellite communications

SÁSusana Gómez Álvarez AMA. Mozo TNTomás Navarro

Key Points

To evaluate whether structured multi-agent debate frameworks and model diversity improve reasoning reliability and decision quality in high-stakes satellite communication tasks.
Compared a single-agent baseline to a multi-agent framework (one moderator and two domain-specialized experts) using 70B-parameter LLMs in homogeneous (Llama-3.3) and heterogeneous (Llama-3.3, DeepSeek-R1, and Qwen-2.5) setups.
Integrated a curated Retrieval-Augmented Generation (RAG) corpus combining academic materials and European Space Agency Nebula portal documentation.
Assessed 213 technical queries across baseline proficiency, strategic reasoning, and executive readiness stages using an LLM-as-a-judge evaluation paradigm.
Single-agent setups achieved higher performance on factual, encyclopedic queries where retrieval alone was sufficient without deliberation.
Multi-agent debate systems surpassed single-agent baselines in complex strategic reasoning tasks across both configurations.
Heterogeneous debate teams achieved the highest performance in executive-readiness scenarios, outperforming single-agent baselines by victory margins of up to 2.75 points on a 10-point scale.

Abstract

Abstract Structured multi-agent debates among Large Language Models (LLMs) have emerged as a powerful paradigm for enhancing reasoning reliability and argumentative coherence. Motivated by the European Space Agency’s (ESA) interest in trustworthy AI for space operations, this study proposes a moderated, domain-adaptive multi-agent debate framework applied to the high-stakes domain of satellite communications (SatCom). Specifically, it assesses (i) the efficacy of structured deliberation against single-agent baselines, and (ii) the impact of model heterogeneity versus homogeneity. A single-agent baseline is compared against a multi-agent framework deploying a moderator and two domain-specialized experts. These systems utilize local 70B-parameter LLMs in homogeneous (Llama-3.3) and heterogeneous (Llama-3.3 + DeepSeek-R1 + Qwen-2.5) configurations, all augmented with a shared, curated Retrieval-Augmented Generation (RAG) corpus combining academic institutional sources and ESA material from the Nebula portal (SatNex V programme). Outputs from 213 technical queries are evaluated via LLM-as-a-judge across three phases: baseline proficiency, strategic reasoning, and executive readiness. Single-agent systems lead in encyclopedic tasks, where retrieval suffices over deliberation. However, both multi-agent configurations outperform in strategic reasoning, with heterogeneous debates achieving superior performance in executive scenarios by victory margins of up to 2.75 points on a 10-point scale. These results validate architectural diversity as a decisive factor in resolving high-complexity technical conflicts. Ultimately, this work delivers a generalizable, fully traceable deliberation framework suitable for real-world, mission-critical environments. Code, prompts, and evaluation data are publicly available at https://github.com/amozo-es/multi-agent-debate/ .

Ask AI

Helpful

Bookmark

View Full Paper