让多智能体系统在运行时自动重组，解决单个智能体过载问题。
Autonomous Topology Mutation: Safe Runtime Restructuring for Multi-Agent LLM Systems with Capability, State, and Shadow Invariants
- 通过监控六项指标检测过载，触发智能体拆分与角色变更。
- 代码任务成功率从3.3%提升至61.7%，隐私数据暴露降至0。
- 保障安全重构，适合高可靠性、高复杂度的智能体系统使用。
多智能体大模型框架通常在启动时固定团队拓扑结构。当某个智能体因任务类型混杂、工具错误累积或请求排队过多而过载时，系统缺乏自我重构能力。本文提出自主拓扑变异（ATM）机制，结合遥测驱动的过载检测与三项安全不变量：能力单调性、状态路由完备性、影子验证优先。ATM监测包含队列深度、上下文抖动、工具错误率、角色熵、重试循环率和跨智能体等待时间的六信号瓶颈指数。当多个连续周期突破预热校准阈值，系统将过载智能体分解为专用子智能体，并将其热替换为协调者角色，保持外部身份一致。状态迁移由隐私级别感知路由控制：每个记忆单元仅传至允许的子集，或记录原因后丢弃。任何候选拓扑在通过影子验证窗口前不接收实时流量。在720次基于DeepSeek-V3的任务运行中，使用确定性工具桩，在四种消融条件和三种工作负载下，因子化分裂使代码任务成功率从3.3%提升至61.7%。完整系统将基于正则表达式的高隐私内存暴露从每任务2.0次降至0.0次，同时维持任务质量。运行时引入的不变量检查对代理主路径的p99延迟影响不足500微秒。一个含真实Python执行的小型在线探针用于外部有效性验证。实现、基准测试工具和日志均已开源。
原文摘要 · Abstract (English)
Multi-agent LLM frameworks typically fix their team topology at boot time. When an individual agent becomes overloaded at runtime, for example by mixing too many action categories, accumulating tool errors, or queueing behind too many calls, the system has no mechanism to restructure itself. We introduce Autonomous Topology Mutation (ATM), a runtime team-mutation mechanism for multi-agent LLM frameworks. ATM combines telemetry-driven overload detection with three safety invariants that gate each structural change: capability monotonicity, state-routing completeness, and shadow-before-live validation. ATM monitors a six-signal Bottleneck Index that includes queue depth, context thrash, tool-error rate, role entropy, retry-loop rate, and cross-agent wait time. When a warmup-calibrated threshold is breached for multiple consecutive ticks, ATM factorises the overloaded agent into specialised sub-agents and hot-swaps the parent into a coordinator role while preserving its external identity. State transfer is controlled by privacy-level-aware routing: each memory atom is routed only to a permitted child set, or explicitly dropped with a logged reason. No candidate topology receives live traffic until it has passed a shadow validation window. On 720 DeepSeek-V3-driven task runs with deterministic tool stubs across four ablation conditions and three workloads, the ATM factoriser split lifts code-task success from 3.3% to 61.7%. The full rail-and-distillation system reduces detected high-privacy memory exposure under a regex classifier from 2.0 to 0.0 events per task while preserving task quality. The runtime rails carrying ATM's invariants add less than 500 microseconds of p99 latency on the agent hot path. A small live-tool probe with real Python execution is included as an external-validity check. The implementation, benchmark harness, and traces are open-sourced.
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。