Autonomous agents powered by large language models (LLMs) have shown impressive capabilities in tool manipulation for complex task-solving. However, existing paradigms such as ReAct rely on sequential reasoning and execution, failing to exploit the inherent parallelism among independent sub-tasks. This sequential bottleneck leads to inefficient tool utilization and suboptimal performance in multi-step reasoning scenarios. We introduce Graph-based Agent Planning (GAP), a novel framework that explicitly models inter-task dependencies through graph-based planning to enable adaptive parallel and serial tool execution. Our approach trains agent foundation models to decompose complex tasks into dependency-aware sub-task graphs, autonomously determining which tools can be executed in parallel and which must follow sequential dependencies. This dependency-aware orchestration achieves substantial improvements in both execution efficiency and task accuracy. To train GAP, we construct a high-quality dataset of graph-based planning traces derived from the Multi-Hop Question Answering (MHQA) benchmark. We employ a two-stage training strategy: supervised fine-tuning (SFT) on the curated dataset, followed by reinforcement learning (RL) with a correctness-based reward function on strategically sampled queries where tool-based reasoning provides maximum value. Experimental results on MHQA datasets demonstrate that GAP significantly outperforms traditional ReAct baselines, particularly on multi-step retrieval tasks, while achieving dramatic improvements in tool invocation efficiency through intelligent parallelization. The project page is available at: https://github.com/WJQ7777/Graph-Agent-Planning.

翻译：基于大语言模型（LLMs）的自主智能体在复杂任务解决的工具操作方面展现出卓越能力。然而，现有范式如ReAct依赖于顺序推理与执行，未能充分利用独立子任务间的内在并行性。这种顺序瓶颈导致工具利用效率低下，在多步推理场景中表现欠佳。本文提出基于图的智能体规划（GAP），一种通过图规划显式建模任务间依赖关系的新型框架，以实现自适应并行与串行工具执行。我们的方法训练智能体基础模型，将复杂任务分解为依赖感知的子任务图，自主决定哪些工具可并行执行、哪些必须遵循顺序依赖。这种依赖感知的编排在执行效率和任务准确性上均取得显著提升。为训练GAP，我们基于多跳问答（MHQA）基准构建了高质量的图规划轨迹数据集，采用两阶段训练策略：首先在精选数据集上进行监督微调（SFT），随后通过强化学习（RL）对策略性采样的查询（其中基于工具的推理能提供最大价值）应用基于正确性的奖励函数。在MHQA数据集上的实验结果表明，GAP显著优于传统ReAct基线方法，尤其在多步检索任务中表现突出，同时通过智能并行化实现了工具调用效率的大幅提升。项目页面详见：https://github.com/WJQ7777/Graph-Agent-Planning。