RKEFino1: A Regulation Knowledge-Enhanced Large Language Model

TL;DR

RKEFino1 is a large language model enhanced with financial regulation knowledge, designed to improve accuracy and compliance in digital regulatory reporting through domain-specific fine-tuning and novel reasoning tasks.

Contribution

It introduces RKEFino1, a regulation knowledge-enhanced LLM for financial reasoning, incorporating domain knowledge from XBRL, CDM, and MOF, and proposing new QA and Numerical NER tasks.

Findings

Demonstrates effectiveness in compliance-critical financial tasks

Shows strong generalization capacity in financial reasoning

Achieved successful fine-tuning with domain-specific knowledge

Abstract

Recent advances in large language models (LLMs) hold great promise for financial applications but introduce critical accuracy and compliance challenges in Digital Regulatory Reporting (DRR). To address these issues, we propose RKEFino1, a regulation knowledge-enhanced financial reasoning model built upon Fino1, fine-tuned with domain knowledge from XBRL, CDM, and MOF. We formulate two QA tasks-knowledge-based and mathematical reasoning-and introduce a novel Numerical NER task covering financial entities in both sentences and tables. Experimental results demonstrate the effectiveness and generalization capacity of RKEFino1 in compliance-critical financial tasks. We have released our model on Hugging Face.