Tools as Continuous Flow for Evolving Agentic Reasoning

TL;DR

FlowAgent introduces a continuous, global planning approach for tool chaining in LLMs, improving robustness and generalization in long-horizon reasoning tasks.

Contribution

It reconceptualizes tool chaining as continuous trajectory generation in semantic space and provides formal bounds on utility convergence.

Findings

FlowAgent outperforms existing methods in robustness and adaptability.

It introduces the first plan-level closed-loop benchmark for agentic reasoning.

Empirical results demonstrate improved long-horizon reasoning performance.

Abstract

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.