FinReporting: An Agentic Workflow for Localized Reporting of Cross-Jurisdiction Financial Disclosures

2026-04-07

AI Summary

This study addresses the challenges of semantic alignment and verification in cross-jurisdictional financial disclosures, which arise from divergences in accounting standards, document formats (e.g., XBRL vs. PDF), and aggregation rules. To overcome these issues, the authors propose an ontology-based structured approach that constructs a unified financial statement ontology and implements a multi-stage, auditable pipeline. Within this framework, large language models (LLMs) are employed not as free-form generators but as constrained validators operating under explicit rule-based guidance and grounded in verifiable evidence, enabling localized information processing. The method innovatively incorporates agent workflows to shift LLMs toward controlled validation, facilitating semantic mapping and anomaly tracing across markets. Experiments on annual reports from the U.S., China, and Japan demonstrate significant improvements in consistency and reliability under heterogeneous disclosure regimes, supported by an interactive platform enabling structured data export.

Abstract

Financial reporting systems increasingly use large language models (LLMs) to extract and summarize corporate disclosures. However, most assume a single-market setting and do not address structural differences across jurisdictions. Variations in accounting taxonomies, tagging infrastructures (e.g., XBRL vs. PDF), and aggregation conventions make cross-jurisdiction reporting a semantic alignment and verification challenge. We present FinReporting, an agentic workflow for localized cross-jurisdiction financial reporting. The system builds a unified canonical ontology over Income Statement, Balance Sheet, and Cash Flow, and decomposes reporting into auditable stages including filing acquisition, extraction, canonical mapping, and anomaly logging. Rather than using LLMs as free-form generators, FinReporting deploys them as constrained verifiers under explicit decision rules and evidence grounding. Evaluated on annual filings from the US, Japan, and China, the system improves consistency and reliability under heterogeneous reporting regimes. We release an interactive demo supporting cross-market inspection and structured export of localized financial statements. Our demo is available at https://huggingface.co/spaces/BoomQ/FinReporting-Demo . The video describing our system is available at https://www.youtube.com/watch?v=f65jdEL31Kk

Problem

• cross-jurisdiction financial reporting
• semantic alignment
• accounting taxonomies
• financial disclosures
• heterogeneous reporting regimes

Innovation

• agentic workflow
• canonical ontology
• cross-jurisdiction financial reporting
• constrained LLM verification
• semantic alignment