arXiv 2602.057942026-02-13cs.AIcs.CEcs.CLcs.LG

FiMI：面向印度金融生态的领域专用语言模型

FiMI: A Domain-Specific Language Model for Indian Finance Ecosystem

National Payments Corporation of India (NPCI)（印度国家支付公司）

Aboli Kathar, Aman Kumar, Anusha Kamath, Araveeti Srujan, Ashish Sharma, Chandra Bhushan, Divya Sorate, Duddu Prasanth Kumar, Evan Acharya, Harsh Sharma, Hrithi… 展开作者

Aboli Kathar, Aman Kumar, Anusha Kamath, Araveeti Srujan, Ashish Sharma, Chandra Bhushan, Divya Sorate, Duddu Prasanth Kumar, Evan Acharya, Harsh Sharma, Hrithik Kadam, Kanishk Singla, Keyur Doshi, Kiran Praveen, Kolisetty Krishna SK, Krishanu Adhikary, Lokesh MPT, Mayurdeep Sonowal, Nadeem Shaikh, Navya Prakash, Nimit Kothari, Nitin Kukreja, Prashant Devadiga, Rakesh Paul, Ratanjeet Pratap Chauhan, Raunak Kalani, Raviraj Joshi, Shamanth MH, Shantanu Pandey, Shubham Soni, Siddharth Dixit, Smriti Jopat, Sunil Patel, Suraj Singh, Suvradip Paul, Tulasi Pilla, Utkarsh Vaidya, Vineeth Nambiar, Vishal Kanvaty, Yatharth Dedhia

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

FiMI是印度国家支付清算公司为数字支付系统开发的领域专用金融语言模型，通过多阶段训练在金融推理和工具调用任务中取得显著性能提升。

摘要：

我们介绍了FiMI（印度金融模型），这是由印度国家支付清算公司（NPCI）为印度数字支付系统开发的领域专用金融语言模型。我们开发了两种模型变体：FiMI Base和FiMI Instruct。FiMI通过多阶段训练流程适配Mistral Small 24B架构，首先在680亿个经过筛选的金融、多语言（英语、印地语、Hinglish）和合成数据上进行连续预训练。随后进行指令微调和领域专用的监督微调，专注于多轮、工具驱动的对话，模拟现实世界的工作流程，如交易纠纷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比Mistral Small 24B Base模型提高了20%，而FiMI Instruct在领域专用工具调用任务上比Mistral Small 24B Instruct模型高出87%。此外，FiMI在保持与类似规模模型在通用基准测试中表现相当的同时，实现了显著的领域优势。

英文摘要：

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.

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。