Multi-agent LLM orchestration incurs synchronization costs scaling as O(n x S x |D|) in agents, steps, and artifact size under naive broadcast -- a regime I term broadcast-induced triply-multiplicative overhead. I argue this pathology is a structural residue of full-state rebroadcast, not an inherent property of multi-agent coordination. The central claim: synchronization cost explosion in LLM multi-agent systems maps with formal precision onto the cache coherence problem in shared-memory multiprocessors, and MESI-protocol invalidation transfers to artifact synchronization under minimal structural modification. I construct the Artifact Coherence System (ACS) and prove the Token Coherence Theorem: lazy invalidation attenuates cost by at least S/(n + W(d_i)) when S > n + W(d_i), converting O(n x S x |D|) to O((n + W) x |D|). A TLA+-verified protocol enforces single-writer safety, monotonic versioning, and bounded staleness across ~2,400 explored states. Simulation across four workload configurations yields token savings of 95.0% +/- 1.3% at V=0.05, 92.3% +/- 1.4% at V=0.10, 88.3% +/- 1.5% at V=0.25, and 84.2% +/- 1.3% at V=0.50 -- each exceeding the theorem's conservative lower bounds. Savings of ~81% persist at V=0.9, contrary to the predicted collapse threshold. Contributions: (1) formal MESI-to-artifact state mapping; (2) Token Coherence Theorem as savings lower bound; (3) TLA+-verified protocol with three proven invariants; (4) characterization of conditional artifact access semantics resolving the always-read objection; (5) reference Python implementation integrating with LangGraph, CrewAI, and AutoGen via thin adapter layers.

翻译：多智能体LLM编排在朴素广播机制下会产生随智能体数量、步数和工件规模以O(n × S × |D|)增长的同步开销——我将此状态称为广播诱导的三重乘法开销。我认为这一病理现象源于全状态重广播的结构性残留，而非多智能体协调的固有特性。核心主张是：LLM多智能体系统中的同步开销爆炸问题在形式化精度上可映射到共享内存多处理器中的缓存一致性问题，且MESI协议的失效机制经最小结构修改即可迁移至工件同步场景。本文构建了工件一致性系统（ACS），并证明令牌一致性定理：当S > n + W(d_i)时，惰性失效机制可将开销降低至少S/(n + W(d_i))，从而将O(n × S × |D|)复杂度转化为O((n + W) × |D|)。经TLA+验证的协议在约2400个探索状态中确保了单写入器安全性、单调版本控制及有限过时性。在四种工作负载配置下的仿真显示：当V=0.05时令牌节省率达95.0% ± 1.3%，V=0.10时为92.3% ± 1.4%，V=0.25时为88.3% ± 1.5%，V=0.50时为84.2% ± 1.3%——各项结果均超越定理的保守下界。在V=0.9时仍保持约81%的节省率，与预测的崩溃阈值相悖。主要贡献包括：（1）形式化的MESI-工件状态映射；（2）作为节省率下界的令牌一致性定理；（3）具备三个可证明不变量的TLA+验证协议；（4）解决“始终读取”异议的条件化工件访问语义表征；（5）通过轻量适配层集成LangGraph、CrewAI和AutoGen的参考Python实现。