arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.16873cs.MAcs.AI
收藏

AdaptOrch:在LLM性能收敛时代的任务自适应多智能体协调

AdaptOrch: Task-Adaptive Multi-Agent Orchestration in the Era of LLM Performance Convergence

Department of Artificial Intelligence, Korea National Open University(人工智能系,韩国国立开放大学)

Geunbin Yu

AI总结:

AdaptOrch通过动态选择多智能体协调拓扑,提升任务性能,证明协调设计优于模型选择。

摘要:

随着大型语言模型来自不同提供商的性能趋于接近,传统的选择单个最佳模型 per 任务的范式逐渐失去效果。我们主张协调拓扑——多个智能体如何协调、并行和合成的结构组成——现在决定了系统层面的性能,而非单个模型的能力。我们提出了AdaptOrch,一个正式的框架,用于任务自适应的多智能体协调,该框架根据任务依赖图和经验得出的领域特征,动态选择四种经典拓扑(并行、顺序、分层和混合)之一。我们的框架引入了三个关键贡献:(1)性能收敛缩放定律,正式化了在何种条件下协调选择会超越模型选择;(2)拓扑路由算法,将任务分解DAG映射到最优协调模式,时间复杂度为O(|V| + |E|);(3)自适应合成协议,具有可证明的终止保证和启发式一致性评分用于并行智能体输出。我们验证了AdaptOrch在编程(SWE-bench)、推理(GPQA)和检索增强生成任务中的表现,证明拓扑感知的协调在使用相同基础模型时,比静态单拓扑基线提高了12-23%。我们的结果确立了协调设计作为首要优化目标的地位,而与模型扩展无关。

英文摘要:

As large language models from diverse providers converge toward comparable benchmark performance, the traditional paradigm of selecting a single best model per task yields diminishing returns. We argue that orchestration topology -- the structural composition of how multiple agents are coordinated, parallelized, and synthesized -- now dominates system-level performance over individual model capability. We present AdaptOrch, a formal framework for task-adaptive multi-agent orchestration that dynamically selects among four canonical topologies (parallel, sequential, hierarchical, and hybrid) based on task dependency graphs and empirically derived domain characteristics. Our framework introduces three key contributions: (1) a Performance Convergence Scaling Law, formalizing conditions under which orchestration selection outweighs model selection; (2) a Topology Routing Algorithm that maps task decomposition DAGs to optimal orchestration patterns in O(|V| + |E|) time; and (3) an Adaptive Synthesis Protocol with provable termination guarantees and heuristic consistency scoring for parallel agent outputs. We validate AdaptOrch across coding (SWE-bench), reasoning (GPQA), and retrieval-augmented generation tasks, demonstrating that topology-aware orchestration achieves 12-23% improvement over static single-topology baselines, even when using identical underlying models. Our results establish orchestration design as a first-class optimization target independent of model scaling.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑