受群体智能启发，构建可自组织协作的分布式推理系统。
SwarmSys: Decentralized Swarm-Inspired Agents for Scalable and Adaptive Reasoning
- 三类角色循环互动：探索者、执行者、验证者协同推进任务。
- 在符号推理等任务中，准确率与推理稳定性显著优于基线模型。
- 无需中心控制，支持动态分配任务，适合复杂长周期推理场景。
大型语言模型代理展现出强大的推理能力，但现有多代理框架常依赖固定角色或中心化控制，限制了长时程推理中的可扩展性与适应性。我们提出 SwarmSys，一种受群体智能启发的闭环分布式多代理推理框架。其协调机制通过探索者、执行者、验证者三类角色的持续迭代交互实现，分别负责探索、执行与验证。为实现可扩展且自适应的合作，系统集成动态代理与事件特征、基于嵌入的概率匹配，以及类信息素强化机制，支持无全局监督下的动态任务分配与自组织收敛。在符号推理、研究综述与科学编程任务中，SwarmSys 均稳定超越基线，提升准确率与推理稳定性。结果表明，群体智能式协调是实现可扩展、鲁棒且自适应多代理推理的有前景范式，提示协调扩展可能与模型扩展同等重要，共同推动大模型智能发展。
原文摘要 · Abstract (English)
Large language model (LLM) agents have shown remarkable reasoning abilities. However, existing multi-agent frameworks often rely on fixed roles or centralized control, limiting scalability and adaptability in long-horizon reasoning. We introduce SwarmSys, a closed-loop framework for distributed multi-agent reasoning inspired by swarm intelligence. Coordination in SwarmSys emerges through iterative interactions among three specialized roles, Explorers, Workers, and Validators, that continuously cycle through exploration, exploitation, and validation. To enable scalable and adaptive collaboration, we integrate adaptive agent and event profiles, embedding-based probabilistic matching, and a pheromone-inspired reinforcement mechanism, supporting dynamic task allocation and self-organizing convergence without global supervision. Across symbolic reasoning, research synthesis, and scientific programming tasks, SwarmSys consistently outperforms baselines, improving both accuracy and reasoning stability. These findings highlight swarm-inspired coordination as a promising paradigm for scalable, robust, and adaptive multi-agent reasoning, suggesting that coordination scaling may rival model scaling in advancing LLM intelligence.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。