DynaSwarm: Dynamically Graph Structure Selection for LLM-based Multi-agent System
This addresses the need for more flexible and efficient multi-agent systems in AI, though it is incremental as it builds on existing RL and fine-tuning methods.
The paper tackles the problem of static collaboration graph structures limiting adaptability and performance in multi-agent systems by proposing DynaSwarm, a dynamic framework that uses reinforcement learning and fine-tuning to optimize graph structures per input sample, resulting in consistent outperformance of state-of-the-art baselines on tasks like question answering, mathematical reasoning, and coding.
Current multi-agent systems (MAS) frameworks often rely on manually designed and static collaboration graph structures, limiting adaptability and performance. To address these limitations, we propose DynaSwarm, a dynamic framework that enhances LLM-based MAS through two key innovations: (1) an actor-critic reinforcement learning (A2C) mechanism to optimize graph structures with improved stability over prior RL methods, and (2) a dynamic graph selector that adaptively chooses the optimal graph structure for each input sample via parameter-efficient LLM fine-tuning. DynaSwarm eliminates the need for rigid, one-fits-all graph architectures, instead leveraging sample-specific idiosyncrasies to dynamically route queries through specialized agent networks. (c) We propose to fine-tune the demonstration retriever to fully exploit the power of in-context learning (ICL). Extensive experiments on question answering, mathematical reasoning, and coding tasks demonstrate that DynaSwarm consistently outperforms state-of-the-art single-agent and MAS baselines across multiple LLM backbones. Our findings highlight the importance of sample-aware structural flexibility in LLM MAS designs.