论文HF Daily Papers · arXiv 2611.05275昨天

Route Anything框架：机器人操作效率提升8.8倍且性能超固定模型

提出通用AI路由框架Route Anything，基于37万条全模态轨迹数据验证。在机器人操作任务中，学习型路由器性能超最佳固定模型1.1分，策略耗时减少24%，平均效率提升8.82倍，为具身智能异构模型编排提供理论与实证支撑。

AI 整理

具身模型
端侧部署

本文由 AI 根据公开内容整理，可能有误，请以原文为准：Route Anything: A Study of Model and Policy Routing for General-Purpose AI Systems（HF Daily Papers · arXiv 2611.05275）

核心内容

针对AGI系统需协调异构模型与工具的痛点，研究提出Route Anything统一框架，建立质量-成本权衡下的最优路由理论基础。团队构建覆盖语言、视觉及机器人操作等25个基准的全模态轨迹矩阵，包含452种候选配置的374,401次评分运行数据，系统性评估了模型与策略路由在通用AI系统中的效能边界。

实验表明，输入级预言机路由比最佳固定候选高5.4分，学习型路由器在16个基准上平均高出0.3分。在机器人操纵任务中，学习路由器性能超越最佳固定模型1.1分，同时策略执行时间减少24%。相比参考模型，路由机制使系统平均效率提升8.82倍，证实了动态编排专用模型在保障具身任务性能的同时可显著降低计算开销。

关键看点

数据集规模：构建含374,401次评分运行的全模态轨迹矩阵，覆盖25个基准与452种候选配置
操作性能增益：学习型路由器在机器人操纵任务中比最佳固定模型高1.1分，策略耗时降24%
系统效率跃升：路由机制使通用AI系统平均效率较参考模型提升8.82倍，验证异构编排价值
理论上限界定：输入级预言机路由比最佳固定候选高5.4分，确立当前路由技术的性能天花板

工程与产业观察

该研究证实具身智能无需依赖单一超大模型，通过路由机制动态调度专用策略可在机器人操作中实现性能与效率双赢。其构建的大规模全模态轨迹矩阵为训练具身路由模型提供了关键数据资产，8.82倍的效率提升指标为边缘侧部署异构专家模型提供了明确的工程优化坐标。

英文摘要 (Abstract)

Artificial general intelligence (AGI) requires a method that can understand, generate, reason, and act across diverse domains and modalities. Rather than relying on a single model to master every capability, a promising path toward AGI now is to orchestrate a heterogeneous ecosystem of specialized models, tools, and policies (candidates) into a unified AI system. At the heart of this vision lies a fundamental question: how can an AI system dynamically select and coordinate the most capable candidate for each task while balancing performance and computational cost? In this work, we present Route Anything, a unified framework for studying model and policy routing across general-purpose AI systems. We develop a unified theoretical foundation that characterizes optimal routing under quality--cost trade-offs and establishes how the information available to a router determines its attainable performance. To systematically investigate routing across diverse capabilities, we first construct an omni-modal trajectory matrix spanning 25 benchmarks across language understanding, image, video, and speech generation, coding and workplace agents, and robotic manipulation, comprising 374,401 scored runs from 452 candidate configurations. Building on this framework, we train and test different routers to investigate the potential of omni-modal model routing, the capabilities of learned routers, and the conditions under which routing improves efficiency. Our results reveal substantial opportunities for capability orchestration: per-input oracle routing outperforms the best fixed candidate by 5.4 points, and learned routers outperform it by 0.3 points across 16 benchmarks, by 2.1 points on video generation, and by 1.1 points on robot manipulation while using 24% less policy time. More importantly, routing raises efficiency by 8.82x on average compared with the reference model.

Hugging Face Daily Papers 热度：86 upvotes 作者：Pengfei Zhou, Yixing Ma, Hexin Wang, Weidong Tang, Zhenglin Wan, Wangbo Zhao, Yang You