路由一切:通用人工智能系统的模型与策略路由研究
Route Anything: A Study of Model and Policy Routing for General-Purpose AI Systems
AI总结:
本研究提出Route Anything框架,构建全模态轨迹矩阵,发现路由可在平衡性能与成本时提升AI系统能力编排效率,平均效率提升8.82倍。
摘要:
通用人工智能(AGI)需要一种能在不同领域和模态间实现理解、生成、推理与行动的方法。当前通往AGI的有前景路径,并非依赖单一模型掌握所有能力,而是将由专用模型、工具和策略(候选体)组成的异构生态系统编排为统一的AI系统。这一愿景的核心问题是:AI系统如何在平衡性能与计算成本的同时,为每个任务动态选择并协调最具能力的候选体?本研究提出Route Anything,这是一个用于研究通用AI系统中模型与策略路由的统一框架。我们构建了统一理论基础,刻画了质量-成本权衡下的最优路由,并确立了路由器可用信息如何决定其可实现性能。为系统研究不同能力下的路由,我们首先构建了全模态轨迹矩阵,涵盖语言理解、图像、视频与语音生成、编码与工作场所智能体、机器人操控等领域的25个基准,包含来自452种候选配置的374401次评分运行。基于该框架,我们训练并测试了不同路由器,以探究全模态模型路由的潜力、学习型路由器的能力,以及路由提升效率的条件。结果显示,能力编排存在巨大机会:每输入的神谕路由比最佳固定候选体高出5.4个点,在16个基准上,学习型路由器比最佳固定候选体高出0.3个点,在视频生成上高出2.1个点,在机器人操控上高出1.1个点,同时策略时间减少24%;更重要的是,与参考模型相比,路由平均提升效率8.82倍。
英文摘要:
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.
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。