用合成数据训练轻量模型，高效准确验证金融事实
FISCAL: Financial Synthetic Claim-document Augmented Learning for Efficient Fact-Checking
- 生成金融领域专用合成数据，增强模型事实核查能力
- 轻量模型精度接近20倍大的模型，超越GPT-3.5 Turbo等同类
- 适合需要高效、可靠金融AI的机构与开发者使用
大型语言模型在金融应用中需兼具事实可靠性与计算效率，但现有系统常产生幻觉且依赖过大模型。本文提出FISCAL（Financial Synthetic Claim-Document Augmented Learning），一种面向金融事实核查的模块化合成数据生成框架。基于FISCAL构建了名为FISCAL-data的合成数据集，并用于训练轻量级验证器MiniCheck-FISCAL。该模型性能超越基线，优于同等规模的GPT-3.5 Turbo及其他开源模型，逼近20倍大的Mixtral-8x22B与Command R+的准确率。在外部数据集FinDVer与Fin-Fact上，其表现媲美GPT-4o与Claude-3.5，优于Gemini-1.5 Flash。结果表明，领域专用合成数据结合高效微调，使紧凑模型实现顶尖的准确性、鲁棒性与可扩展性，适用于实际金融AI部署。数据集与代码已开源。
原文摘要 · Abstract (English)
Financial applications of large language models (LLMs) require factual reliability and computational efficiency, yet current systems often hallucinate details and depend on prohibitively large models. We propose FISCAL (Financial Synthetic Claim-Document Augmented Learning), a modular framework for generating synthetic data tailored to financial fact-checking. Using FISCAL, we generate a dataset called FISCAL-data and use it to train MiniCheck-FISCAL, a lightweight verifier for numerical financial claims. MiniCheck-FISCAL outperforms its baseline, surpasses GPT-3.5 Turbo and other open-source peers of similar size, and approaches the accuracy of much larger systems (20x), such as Mixtral-8x22B and Command R+. On external datasets FinDVer and Fin-Fact, it rivals GPT-4o and Claude-3.5 while outperforming Gemini-1.5 Flash. These results show that domain-specific synthetic data, combined with efficient fine-tuning, enables compact models to achieve state-of-the-art accuracy, robustness, and scalability for practical financial AI. The dataset and scripts are available in the project repository (link provided in the paper).
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。