Can Lessons From Human Teams Be Applied to Multi-Agent Systems? The Role of Structure, Diversity, and Interaction Dynamics
This work addresses the need for better team performance in multi-agent systems for AI researchers, though it is incremental as it applies existing human team science to a new AI context.
The study tackled the problem of understanding team dynamics in multi-agent systems by applying human team science concepts to LLM-powered agents, finding that flat teams outperformed hierarchical ones on commonsense and social reasoning tasks, with diversity showing nuanced effects and agents displaying overconfidence and coordination challenges.
Multi-Agent Systems (MAS) with Large Language Model (LLM)-powered agents are gaining attention, yet fewer studies explore their team dynamics. Inspired by human team science, we propose a multi-agent framework to examine core aspects of team science: structure, diversity, and interaction dynamics. We evaluate team performance across four tasks: CommonsenseQA, StrategyQA, Social IQa, and Latent Implicit Hate, spanning commonsense and social reasoning. Our results show that flat teams tend to perform better than hierarchical ones, while diversity has a nuanced impact. Interviews suggest agents are overconfident about their team performance, yet post-task reflections reveal both appreciation for collaboration and challenges in integration, including limited conversational coordination.