Multi-Agent Large Language Models for Conversational Task-Solving
Jonas Becker
TL;DR
The paper addresses the limitations of single LLMs by proposing Multi-Agent LLMs for conversational task-solving. It introduces MALLM, a modular framework and taxonomy that decouples agents, discussion paradigms, and decision-making to study how multiple expert personas interact in dialogue. Empirical results show that multi-agent LLMs enhance complex reasoning and ethical alignment, but can underperform on basic tasks due to problem drift and alignment hazards; centralized, information-sharing paradigms can mitigate some safety concerns. The work highlights the tradeoffs between discussion length, task complexity, and agent composition, and offers guidelines and directions for safer, more efficient multi-agent AI systems. Overall, it provides a rigorous foundation for evaluating how multi-agent interactions influence performance across generative and QA tasks and points to future research on safety, fairness, and scalable deployment.
Abstract
In an era where single large language models have dominated the landscape of artificial intelligence for years, multi-agent systems arise as new protagonists in conversational task-solving. While previous studies have showcased their potential in reasoning tasks and creative endeavors, an analysis of their limitations concerning the conversational paradigms and the impact of individual agents is missing. It remains unascertained how multi-agent discussions perform across tasks of varying complexity and how the structure of these conversations influences the process. To fill that gap, this work systematically evaluates multi-agent systems across various discussion paradigms, assessing their strengths and weaknesses in both generative tasks and question-answering tasks. Alongside the experiments, I propose a taxonomy of 20 multi-agent research studies from 2022 to 2024, followed by the introduction of a framework for deploying multi-agent LLMs in conversational task-solving. I demonstrate that while multi-agent systems excel in complex reasoning tasks, outperforming a single model by leveraging expert personas, they fail on basic tasks. Concretely, I identify three challenges that arise: 1) While longer discussions enhance reasoning, agents fail to maintain conformity to strict task requirements, which leads to problem drift, making shorter conversations more effective for basic tasks. 2) Prolonged discussions risk alignment collapse, raising new safety concerns for these systems. 3) I showcase discussion monopolization through long generations, posing the problem of fairness in decision-making for tasks like summarization. This work uncovers both the potential and challenges that arise with multi-agent interaction and varying conversational paradigms, providing insights into how future research could improve the efficiency, performance, and safety of multi-agent LLMs.