arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2503.14432cs.CLcs.AIcs.LG
收藏

PLAY2PROMPT:通过工具试玩实现大语言模型智能体的零样本工具指令优化

PLAY2PROMPT: Zero-shot Tool Instruction Optimization for LLM Agents via Tool Play

Massachusetts Institute of Technology(麻省理工学院);MIT-IBM Watson AI Lab(MIT-IBM沃森人工智能实验室)

Wei Fang, Yang Zhang, Kaizhi Qian, James Glass, Yada Zhu

AI总结:

针对LLM零样本工具使用时文档不足或有噪、现有方案依赖人工或标注数据的问题,提出PLAY2PROMPT框架,通过自动试玩工具探索行为、生成文档和示例,无需标注数据即可显著提升各类模型的零样本工具使用性能。

摘要:

大语言模型(LLM)正日益与各类专用外部工具集成,但许多任务要求在工具文档极少或存在噪声的情况下实现零样本工具使用。现有方案依赖人工重写或带标注数据进行验证,无法适用于真正的零样本场景。为解决这些挑战,我们提出PLAY2PROMPT,这是一个自动化框架,能系统性地对每个工具进行“试玩”以探索其输入输出行为。通过这一迭代试错过程,PLAY2PROMPT无需任何标注数据即可完善工具文档并生成使用示例。这些示例不仅能指导LLM推理,还可作为验证数据进一步提升工具利用效率。在真实世界任务上的大量实验表明,PLAY2PROMPT在开源和闭源模型上均显著提升了零样本工具使用性能,为领域特定工具集成提供了可扩展且有效的解决方案。

英文摘要:

Large language models (LLMs) are increasingly integrated with specialized external tools, yet many tasks demand zero-shot tool usage with minimal or noisy documentation. Existing solutions rely on manual rewriting or labeled data for validation, making them inapplicable in true zero-shot settings. To address these challenges, we propose PLAY2PROMPT, an automated framework that systematically "plays" with each tool to explore its input-output behaviors. Through this iterative trial-and-error process, PLAY2PROMPT refines tool documentation and generates usage examples without any labeled data. These examples not only guide LLM inference but also serve as validation to further enhance tool utilization. Extensive experiments on real-world tasks demonstrate that PLAY2PROMPT significantly improves zero-shot tool performance across both open and closed models, offering a scalable and effective solution for domain-specific tool integration.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑