让大模型分步规划工具检索，解决复杂任务匹配难题。
Beyond Single-Shot: Multi-step Tool Retrieval via Query Planning
- 将复杂指令拆解为子任务，动态生成查询逐层检索
- 在多个工具库上实现领先性能，零样本泛化能力更强
- 适合构建高阶智能体的开发者或研究者使用
基于大型语言模型的智能体在处理海量动态工具库时依赖高效检索，但传统单次匹配的密集检索方法在应对复杂请求时表现不佳。主要问题在于用户目标与技术文档间的语义断层，以及固定长度嵌入难以建模工具组合的复杂性。为此，我们提出TOOLQP——一种轻量级框架，将检索建模为迭代查询规划过程。不同于单次匹配，TOOLQP将指令分解为子任务，并动态生成查询与检索器交互，通过聚焦具体组合所需子任务有效弥合语义差距。我们采用合成查询轨迹训练，并通过可验证奖励的强化学习（RLVR）进行优化。实验表明，TOOLQP达到当前最优性能，展现出卓越的零样本泛化能力、对多种检索器的鲁棒性，以及在下游智能体执行中的显著提升。
原文摘要 · Abstract (English)
LLM agents operating over massive, dynamic tool libraries rely on effective retrieval, yet standard single-shot dense retrievers struggle with complex requests. These failures primarily stem from the disconnect between abstract user goals and technical documentation, and the limited capacity of fixed-size embeddings to model combinatorial tool compositions. To address these challenges, we propose TOOLQP, a lightweight framework that models retrieval as iterative query planning. Instead of single-shot matching, TOOLQP decomposes instructions into sub-tasks and dynamically generates queries to interact with the retriever, effectively bridging the semantic gap by targeting the specific sub-tasks required for composition. We train TOOLQP using synthetic query trajectories followed by optimization via Reinforcement Learning with Verifiable Rewards (RLVR). Experiments demonstrate that TOOLQP achieves state-of-the-art performance, exhibiting superior zero-shot generalization, robustness across diverse retrievers, and significant improvements in downstream agentic execution.
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。