Abstract
Single-agent LLMs hit hard limits--finite context, role overload, and brittle domain transfer. Conventional multi-agent fixes soften those edges yet expose fresh pains: ill-posed decompositions, fuzzy contracts, and verification overhead that blunts the gains. We therefore present Know-The-Ropes (KtR), a framework that converts domain priors into an algorithmic blueprint hierarchy, in which tasks are recursively split into typed, controller-mediated subtasks, each solved zero-shot or with the lightest viable boost (e.g., chain-of-thought, micro-tune, self-check). Grounded in the No-Free-Lunch theorem, KtR trades the chase for a universal prompt for disciplined decomposition. On the Knapsack problem (3-8 items), three GPT-4o-mini agents raise accuracy from 3% zero-shot to 95% on size-5 instances after patching a single bottleneck agent. On the tougher Task-Assignment problem (6-15 jobs), a six-agent o3-mini blueprint hits 100% up to size 10 and 84% on sizes 13-15, versus 11% zero-shot. Algorithm-aware decomposition plus targeted augmentation thus turns modest models into reliable collaborators--no ever-larger monoliths required.
Abstract (translated)
单一代理LLM面临硬性限制——有限的上下文、角色过载和脆弱的知识领域转移。传统多代理解决方案虽然减轻了这些问题，但也暴露出新的问题：不恰当的任务分解、模糊不清的合作协议以及验证成本高昂，削弱了改进效果。因此，我们提出了一种名为“掌握诀窍”（Know-The-Ropes, KtR）的框架，该框架将领域的先验知识转化为算法蓝图层级结构，在这种结构中，任务被递归地拆分为有类型的、由控制器中介的子任务，每个子任务要么直接解决，要么通过最轻量级的方法进行增强（例如：思维链推理、微调或自我检查）。基于“没有免费午餐”的定理，KtR放弃了寻找通用提示符的努力，转而强调有条不紊的任务分解。 在背包问题（3-8个物品）上，使用三个GPT-4o-mini代理，在补全单一瓶颈代理后，从零样本的3%准确率提高到大小为5的情况下的95%。对于更具挑战性的任务分配问题（6-15项工作），一个由六个o3-mini蓝图组成的系统在规模达到10时能够实现100%的正确率，并且在规模13-15时也能保持84%的准确度，相比之下零样本情况下的准确率为11%。 通过算法意识的任务分解加上有针对性的增强，这种框架使中等大小的模型成为可靠的合作伙伴——无需构建越来越大、越来越复杂的单一代理系统。
URL
https://arxiv.org/abs/2505.16979
