AI 中文总结
该研究提出MANTA框架，使多智能体系统通信拓扑可在推理阶段自演进，在五类基准测试中平均得分74.0，较最强基线高5.8个百分点，验证了推理阶段自改进可延伸至协作架构。
AI 中文摘要
基于大语言模型的多智能体系统通过任务分解、智能体专业化、信息交换和中间验证提升复杂问题解决能力，但现有系统通常将通信拓扑视为固定设计选择或离线优化目标。本文提出MANTA，即多智能体网络拓扑自适应框架，使通信结构能在推理阶段自演进。执行前，MANTA基于先验结构经验初始化任务条件拓扑；部署时，它监控协作轨迹，当当前组织不足时应用有界结构更新，这些更新可修改智能体角色、通信链路、执行顺序、信息可见性和验证路径，同时保留任务接口和智能体预算。我们在五个涵盖信息检索、工具使用、规划、工作流执行和数学推理的基准测试中，将MANTA与代表性单智能体和多智能体基线对比，MANTA取得74.0的最高平均得分，比最强基线高出5.8个百分点，并在PlanCraft上获得最佳结果。这些结果表明，推理阶段的自改进可延伸至协作架构本身。
英文摘要
Large language model-based multi-agent systems improve complex problem solving through task decomposition, agent specialization, information exchange, and intermediate validation. However, existing systems typically treat communication topology as a fixed design choice or an offline optimization target. We introduce MANTA, a framework for Multi-Agent Network Topology Adaptation that enables communication structures to self-evolve at inference time. Before execution, MANTA initializes a task-conditioned topology from prior structural experience. During deployment, it monitors collaboration traces and applies bounded structural updates when the current organization becomes insufficient. These updates can modify agent roles, communication links, execution order, information visibility, and validation pathways while preserving the task interface and agent budget. We evaluate MANTA against representative single-agent and multi-agent baselines on five benchmarks spanning information seeking, tool use, planning, workflow execution, and mathematical reasoning. MANTA achieves the highest average score of 74.0, outperforming the strongest baseline by 5.8 percentage points and obtaining the best result on PlanCraft. These results show that inference-time self-improvement can extend to the architecture of collaboration itself.
