Financial Close JSON Consistency Evaluation Benchmark

#JSON consistency verification #financial discrepancy detection #evidence location identification #financial close auditing #financial reconciliation #accounting data quality assurance

3320
0.4GB
JSON
CATL
MOBIUSI INC

Updated:2026-11-03

AI Analysis & Value Prop

This benchmark contains JSON vouchers, account balances, trial balance data, and financial statements from enterprise financial close workflows, with inconsistencies and supporting locations annotated across records. It focuses on checks between the general ledger and financial statements, debit and credit balances, and accounting-period data, enabling evaluation of a model's ability to detect and localize financial data inconsistencies. It is suitable for developing and automatically evaluating consistency-checking capabilities in enterprise financial management software.

Dataset Insights

Sample Examples

{}

Technical Specifications

Field	Type	Description
file_name	string	File name
findings	array	Detailed inconsistency findings, including references to affected records and supporting JSON locations.
source_json	string	The original JSON payload containing financial-close vouchers, account balances, trial balance data, and financial statements, serialized as text.
is_consistent	boolean	Indicates whether the supplied financial records are consistent across the evaluated checks.
issue_categories	array	The categories of inconsistencies identified in the source data.
detected_issue_count	integer	The total number of distinct inconsistencies identified in the source data.

Compliance Statement

Authorization Type	Proprietary - Commercial AI Training License (No Redistribution)
Commercial Use	Requires exclusive subscription or authorization contract (monthly or per-invocation charging)
Privacy and Anonymization	No PII, no real company names, simulated scenarios follow industry standards
Compliance System	Compliant with China's Data Security Law / EU GDPR / supports enterprise data access logs

Frequently Asked Questions

What financial data does the benchmark contain?: It includes JSON vouchers, account balances, trial balance data, and financial statements related to financial close, along with annotated inconsistencies and supporting locations.

How are consistency issues annotated?: Each sample provides an overall consistency status, the number and categories of detected issues, and detailed findings that reference affected records and relevant JSON locations.

Which evaluation tasks does it support?: It can evaluate checks between the general ledger and financial statements, debit and credit balance validation, and accounting-period consistency. It also supports testing discrepancy detection and localization in financial software.

How can I access the data, and what are the licensing terms?: Refer to the product page for available access methods, data specifications, and license terms. Confirm the applicable rights before commercial deployment or redistribution.

Can't find the data you need?

Post a request and let data providers reach out to you.

Get this Dataset

Verified for Enterprise Use

Cite this Work

@dataset{Mobiusi2026,
  title={Financial Close JSON Consistency Evaluation Benchmark},
  author={MOBIUSI INC},
  year={2026},
  url={https://www.mobiusi.com/datasets/4603d33e63cd4fe4cef12a76f4941b80},
  urldate={2026-10-22},
  keywords={financial close data,JSON consistency benchmark,financial data validation,trial balance checks,general ledger reconciliation,debit credit validation,financial anomaly detection,accounting consistency benchmark},
  version={1.0}
}

Using this in research? Please cite us.

placeholder

Scene

Domain

Modality

Task