用虚拟工具合成数据，让小模型超越GPT-4o的多跳搜索能力
ToolForge: A Data Synthesis Pipeline for Multi-Hop Search without Real-World APIs
- 基于问答-上下文-答案三元组生成多跳搜索数据
- 80亿参数模型在合成数据上表现超GPT-4o
- 无需真实API，支持多跳推理与自我反思
训练大模型调用工具并利用检索信息需要高质量、多样化的数据。然而，现有合成数据生成管道通常依赖数万次真实API调用以提升泛化能力，成本高昂且缺乏多跳推理与自我反思能力。为此，我们提出ToolForge，一种自动化合成框架，仅通过构建少量虚拟工具即可实现强现实工具调用性能，无需真实API调用。ToolForge利用（问题，黄金上下文，答案）三元组，专门设计用于多跳搜索场景的数据合成，并通过多跳推理与自我反思机制进一步丰富数据。为确保数据质量，采用多层验证框架，结合规则与模型评估。实证结果表明，仅使用80亿参数的模型，在合成数据上训练后，在多个基准测试中超越GPT-4o。代码与数据集已公开于https://github.com/Buycar-arb/ToolForge。
原文摘要 · Abstract (English)
Training LLMs to invoke tools and leverage retrieved information necessitates high-quality, diverse data. However, existing pipelines for synthetic data generation often rely on tens of thousands of real API calls to enhance generalization, incurring prohibitive costs while lacking multi-hop reasoning and self-reflection. To address these limitations, we introduce ToolForge, an automated synthesis framework that achieves strong real-world tool-calling performance by constructing only a small number of virtual tools, eliminating the need for real API calls. ToolForge leverages a (question, golden context, answer) triple to synthesize large-scale tool-learning data specifically designed for multi-hop search scenarios, further enriching the generated data through multi-hop reasoning and self-reflection mechanisms. To ensure data fidelity, we employ a Multi-Layer Validation Framework that integrates both rule-based and model-based assessments. Empirical results show that a model with only 8B parameters, when trained on our synthesized data, outperforms GPT-4o on multiple benchmarks. Our code and dataset are publicly available at https://github.com/Buycar-arb/ToolForge .
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。