SAP通过状态引导和论证溯源生成高质量多轮工具使用数据
SAP: State-Guided Data Synthesis with Argument Provenance for Multi-Turn Tool Use
- 用状态引导和论证溯源约束生成带长程依赖的工具调用轨迹
- 基于SAP数据训练的40亿参数模型在多个基准上媲美更大模型
- 适合研究智能体、多轮工具调用与数据合成的开发者
高质量的多轮工具使用数据对训练智能体模型至关重要，但现有数据合成方法常忽略关键的论据级依赖关系，导致即使模型选对工具，仍因填充虚构、过时或弱相关参数而失败。为此，我们提出状态引导的数据合成方法（SAP），结合状态引导、工具参数溯源约束与回合级验证，高效构建具有长程依赖且高准确性的工具使用轨迹。利用SAP生成的数据，我们构建了SAP-4B模型，在多个基准测试中表现优异，甚至可媲美更大规模模型。代码、合成数据及训练权重已开源。
原文摘要 · Abstract (English)
High-quality multi-turn tool-use data is essential for training agentic models, yet existing data synthesis methods often underrepresent the argument-level dependencies that are critical to long-horizon tool use. As a result, even when a model selects the correct tool, task execution may still fail because the model fills tool arguments with fabricated, stale, or weakly grounded values. To address this problem, we propose \textbf{State-Guided Data Synthesis with Argument Provenance (SAP)}. SAP combines state guidance, tool-argument provenance constraints, and turn-level validation to efficiently construct tool-use trajectories with long-range dependencies and high accuracy. Using data generated by SAP, we build SAP-4B, which is highly competitive even when compared with much larger models across multiple benchmarks. Source code, synthesized data, and trained weights are available at https://github.com/Zichen1024/SAP.
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。