发表机构
eCampus University; SMARTEST Research Center; V. N. Karazin Kharkiv National University; University of Macerata（埃坎帕斯大学; SMARTEST研究中心; V.N.卡拉津哈尔科夫国立大学; 马切拉塔大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对扁平多智能体系统，构建了基于宽度、内存、延迟的定量资源模型，反驳了其存在不可约因果下限的结论，明确资源不可替代性并给出设计规则。
AI 中文摘要
在可扩展多智能体系统（从机器人群到大型语言模型（LLM）智能体集合）的设计中，一个反复出现的问题是：仅增加智能体数量能否克服性能极限，还是需要具有质的“更深层次”组织？一篇近期预印本指出，扁平同构多智能体系统在可实现误差上面临与种群规模无关的不可约“因果下限”，仅可通过分层（嵌套循环）组织消除。我们使用具有精确可计算最优解的受控干扰抑制测试平台，证明该结论过于绝对，并提出基于三种资源的定量资源模型：种群宽度$N$、单智能体内部模型内存$d$以及观测延迟$\tau$。我们确立三项结论：（i）可实现下限由单智能体内部模型内容而非架构层级决定：在单智能体内存相等时，携带匹配干扰内部模型的扁平同构群体，性能可媲美或优于设计的双循环层级——因此时间深度可以是动态的（循环记忆），而非架构的（嵌套）；（ii）三种资源不可相互替代；我们在显式宽度×内存图上绘制了交换率和硬性非交换边界，包括严格的总状态预算相等比较；（iii）剩余下限由观测延迟和环境在该时间范围内的不可预测性决定，我们通过最优控制器验证了这一点。我们量化了用在线学习替代干扰谱先验知识的代价，提供了针对轻度有界非线性和空间扩展装置的初步鲁棒性检查，并提炼出四条面向从业者的设计规则。
英文摘要
A recurring question in the design of scalable multi-agent systems -- from robot swarms to collectives of large-language-model (LLM) agents -- is whether adding more agents can, on its own, overcome performance limits, or whether a qualitatively \emph{deeper} organization is required. A recent preprint argues that flat, homogeneous multi-agent systems face an irreducible, population-independent ``causal floor'' on achievable error, removable only by hierarchical (nested-loop) organization. Using a controlled disturbance-rejection testbed with an exactly computable optimum, we show this conclusion is too strong and replace it with a quantitative resource model built on three resources: population \emph{width} $N$, per-agent internal-model \emph{memory} $d$, and prediction across the observation \emph{delay} $τ$. We establish three claims. (i) The achievable floor is governed not by architectural hierarchy but by per-agent internal-model content: a flat, homogeneous swarm whose agents carry a matched internal model of the disturbance matches or beats a designed two-loop hierarchy at equal per-agent memory -- so temporal depth can be dynamical (recurrent memory), not architectural (nesting). (ii) The three resources are \emph{not mutually interchangeable}; we chart the exchange rates and the hard non-exchange boundaries on an explicit width$\times$memory map, including a strict equal-total-state-budget comparison. (iii) A residual floor is set by the observation delay and the environment's unpredictability over that horizon, which we verify against the optimal controller. We quantify the price of replacing oracle knowledge of the disturbance spectrum with online learning, provide a preliminary robustness check against a mild bounded nonlinearity and a spatially-extended plant, and distill four design rules for practitioners.
