渐进式工具检索改进规划

摘要

大型语言模型（LLMs）越来越多地用于复杂的多步规划任务中，其中工具检索（TR）步骤对于取得成功结果至关重要。工具检索的两种主要方法是单步检索，利用完整查询，和使用任务分解（TD）的顺序检索，其中完整查询被分割成离散的原子子任务。虽然单步检索缺乏处理“工具间依赖”的灵活性，但TD方法需要保持“子任务-工具原子性对齐”，因为工具箱可以动态演变。为了解决这些限制，我们引入了渐进式工具检索以改进规划（ProTIP）框架。ProTIP是一个基于轻量级对比学习的框架，可以在不需要显式子任务标签的情况下隐式执行TD，同时保持子任务-工具的原子性。在ToolBench数据集上，ProTIP的表现远远优于基于ChatGPT任务分解的方法，TR的Recall@K=10提高了24％，规划生成的工具准确性提高了41％。

English

Large language models (LLMs) are increasingly employed for complex multi-step planning tasks, where the tool retrieval (TR) step is crucial for achieving successful outcomes. Two prevalent approaches for TR are single-step retrieval, which utilizes the complete query, and sequential retrieval using task decomposition (TD), where a full query is segmented into discrete atomic subtasks. While single-step retrieval lacks the flexibility to handle "inter-tool dependency," the TD approach necessitates maintaining "subtask-tool atomicity alignment," as the toolbox can evolve dynamically. To address these limitations, we introduce the Progressive Tool retrieval to Improve Planning (ProTIP) framework. ProTIP is a lightweight, contrastive learning-based framework that implicitly performs TD without the explicit requirement of subtask labels, while simultaneously maintaining subtask-tool atomicity. On the ToolBench dataset, ProTIP outperforms the ChatGPT task decomposition-based approach by a remarkable margin, achieving a 24% improvement in Recall@K=10 for TR and a 41% enhancement in tool accuracy for plan generation.

渐进式工具检索改进规划

ProTIP: Progressive Tool Retrieval Improves Planning

摘要

Support