FiMI是印度支付系统专用的金融大模型，提升本地金融任务表现。
FiMI: A Domain-Specific Language Model for Indian Finance Ecosystem
- 基于Mistral Small架构，用680亿金融语料分阶段训练
- 在金融推理任务上比基线模型提升20%，工具调用准确率高87%
- 专为印度多语言支付场景设计，适合金融从业者与开发者
我们介绍由印度国家支付公司（NPCI）开发的领域专用金融语言模型FiMI（Finance Model for India），用于支持印度数字支付系统。该模型包含两个版本：FiMI Base 和 FiMI Instruct。FiMI通过多阶段训练流程对Mistral Small 24B架构进行适配，首先在680亿条经过筛选的金融、多语言（英语、印地语、印地英混杂语）及合成数据上进行连续预训练。随后进行指令微调和领域特定的监督微调，聚焦于多轮、工具驱动的对话，模拟真实业务流程如交易争议处理和授权生命周期管理。评估显示，FiMI Base在金融推理基准测试中相比Mistral Small 24B Base模型提升20%；而FiMI Instruct在领域特定工具调用任务中，相较Mistral Small 24B Instruct模型提升87%。此外，FiMI在取得显著领域性能提升的同时，保持了与同规模模型相当的一般基准表现。
原文摘要 · Abstract (English)
We present FiMI (Finance Model for India), a domain-specialized financial language model developed by National Payments Corporation of India (NPCI) for Indian digital payment systems. We develop two model variants: FiMI Base and FiMI Instruct. FiMI adapts the Mistral Small 24B architecture through a multi-stage training pipeline, beginning with continuous pre-training on 68 Billion tokens of curated financial, multilingual (English, Hindi, Hinglish), and synthetic data. This is followed by instruction fine-tuning and domain-specific supervised fine-tuning focused on multi-turn, tool-driven conversations that model real-world workflows, such as transaction disputes and mandate lifecycle management. Evaluations reveal that FiMI Base achieves a 20\% improvement over the Mistral Small 24B Base model on finance reasoning benchmark, while FiMI Instruct outperforms the Mistral Small 24B Instruct model by 87\% on domain-specific tool-calling. Moreover, FiMI achieves these significant domain gains while maintaining comparable performance to models of similar size on general benchmarks.
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。