arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.20426cs.AI
收藏

学习重写工具描述以实现可靠的LLM-代理工具使用

Learning to Rewrite Tool Descriptions for Reliable LLM-Agent Tool Use

Intuit AI Research(Intuit AI研究院)

Ruocheng Guo, Kaiwen Dong, Xiang Gao, Kamalika Das

AI总结:

本文提出Trace-Free+框架,通过渐进式转移监督从轨迹丰富设置到轨迹自由部署,提升代理在大规模工具库中的鲁棒性与泛化能力,同时提高查询成功率。

摘要:

尽管大多数改进LLM基于工具使用代理的努力集中在代理本身,通过更大模型、更好提示或微调,代理性能因所消耗工具接口的质量而趋于平缓。工具描述通常为人类开发者编写,容忍模糊性,而代理无法解决,尤其是候选工具数量增加时。现有方法改进工具接口(1)需要为每个进入目录的API重新运行多阶段流程,(2)通常独立优化每个工具,限制了可扩展性和对未见工具的泛化能力。我们提出Trace-Free+,一种课程学习框架,逐步将监督从轨迹丰富设置转移到轨迹自由部署,鼓励模型内化有效工具描述的可重用模式。为支持此方法,我们构建了一个大规模高质量工具接口数据集,通过系统化的数据合成工作流程从真实世界API中获得。在广泛采用的基准测试中,Trace-Free+在工具库扩展到150+候选时提升了鲁棒性,在扩展实验中减少了准确性下降29.23%,并在StableToolBench上提高了平均查询级成功率60.89%。该方法在不同领域间泛化无需重新训练,并在代理微调上提供互补增益。

英文摘要:

While most efforts to improve LLM-based tool-using agents focus on the agent itself - through larger models, better prompting, or fine-tuning - agent performance increasingly plateaus due to the quality of the tool interfaces these agents consume. Tool descriptions are often written for human developers and tolerate ambiguity that agents cannot resolve, particularly as the number of candidate tools grows. Existing approaches to improving tool interfaces (1) require re-running a multi-stage per-tool pipeline - synthesizing queries, executing an agent to collect trajectories, annotating trajectories, and prompting a strong LLM multiple times - for every API that enters the catalog, and (2) typically optimize each tool independently, limiting scalability and generalization to unseen tools. We propose Trace-Free+, a curriculum learning framework that progressively transfers supervision from trace-rich settings to trace-free deployment, encouraging the model to internalize reusable patterns of what makes a tool description effective. To support this approach, we construct a large-scale dataset of high-quality tool interfaces derived from real-world APIs through a principled data synthesis workflow. Experiments on widely adopted benchmarks show that Trace-Free+ improves robustness as tool catalogs scale to 150+ candidates - in scaling experiments, reducing accuracy degradation by 29.23% and improving average query-level success by 60.89% on StableToolBench - generalizes across domains without retraining, and provides complementary gains on top of agent fine-tuning.

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。

↑