让工具调用像流水一样连续演进，提升长程推理的稳定性和泛化能力。
Tools as Continuous Flow for Evolving Agentic Reasoning
- 将工具链视为语义空间中的连续轨迹生成，而非离散步骤
- 在长序列任务中误差更小，对未见过工具的适应性更强
- 适合需要持续决策与动态调整的智能体系统
大型语言模型在协调工具完成推理任务方面展现出卓越能力。然而，现有方法依赖于缺乏全局视角的分步范式，导致长时程任务中错误累积，并限制对未见工具的泛化能力。为此，我们提出工具连续流机制（FlowAgent），将工具链重构为语义空间中的连续轨迹生成。为系统评估该范式，我们首次引入面向动态真实环境的计划级闭环评测基准。FlowAgent利用条件流匹配生成连续潜在轨迹，提供全局规划视角，确保工具执行的一致性与鲁棒性。理论上，我们建立了效用收敛的严格边界，并证明连续形式可从根本上保障泛化能力与误差衰减。实验表明，FlowAgent在长时程推理任务中表现出更优的鲁棒性与适应性。
原文摘要 · Abstract (English)
Large Language Models (LLMs) have demonstrated remarkable capabilities in orchestrating tools for reasoning tasks. However, existing methods rely on a step-wise paradigm that lacks a global perspective, which causes error accumulation over long horizons and restricts generalization to unseen tools. To overcome these limitations, we propose Tools as Continuous Flow for Evolving Agentic Reasoning (FlowAgent), which reconceptualizes tool chaining as continuous trajectory generation within a semantic space. To systematically evaluate this paradigm, we introduce the first plan-level closed-loop benchmark dedicated to plan-level agentic reasoning in dynamic real-world environments. Specifically, the proposed FlowAgent leverages conditional flow matching to generate continuous latent trajectories, providing a global planning perspective to ensure coherent and robust tool execution. Theoretically, we establish formal bounds on utility convergence and prove that our continuous formulation fundamentally guarantees robust generalization and error attenuation. Empirical evaluations show that FlowAgent achieves superior robustness and adaptability in long-horizon reasoning tasks.
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。