让AI团队自动分工，又准又省，效率提升近10倍。
Uno-Orchestra: Parsimonious Agent Routing via Selective Delegation
- 统一学习任务分解与分配，端到端优化
- 在13个基准上达77.0%准确率，比最强基线高16%
- 推理成本降低约90%，适合资源受限场景
大型语言模型多智能体系统通常依赖固定调度策略，要么采用扁平化的查询路由，要么依赖人工设计的任务分解，导致分解深度、工作者选择和推理预算无法在统一目标下协同优化。本文提出Uno-Orchestra，一种统一的调度策略，通过有选择地分解任务并将每个子任务分配给合适的（模型，原语）组合对，两个决策均基于真实工作者交互数据生成的强化学习轨迹联合学习。在涵盖数学、代码、知识、长上下文和代理工具使用等13个基准的评测中，相较于22个基线，Uno-Orchestra在宏平均pass@1上达到77.0%，较最强工作流基线提升约16%，同时每查询成本降低约一个数量级，显著推进了选择性委派的准确率-效率前沿。
原文摘要 · Abstract (English)
Large language model (LLM) multi-agent systems typically rely on rigid orchestration, committing either to flat per-query routing or to hand-engineered task decomposition, so decomposition depth, worker choice, and inference budget are not jointly optimized under one objective. We introduce Uno-Orchestra, a unified orchestration policy that selectively decomposes a task and dispatches each subtask to an admissible (model, primitive) pair, with both decisions learned together from curated RL trajectories grounded in real worker interactions. Against 22 baselines on a 13-benchmark suite spanning math, code, knowledge, long-context, and agentic tool-use, Uno-Orchestra reaches 77.0% macro pass@1, roughly 16% above the strongest workflow baseline, at roughly an order of magnitude lower per-query cost, advancing the accuracy-efficiency frontier of selective delegation.
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。