AuditFlow：面向结构化财务报告验证的可执行符号环境

View PDF HTML (experimental)

Abstract:Structured financial reporting verification requires connecting reported facts to accounting constraints, reconstructing expected values, and executing numerical checks. We study this problem through XBRL numerical consistency verification and introduce \textsc{AuditFlow}, a graph-grounded multi-agent framework built around an executable symbolic environment. The environment links a reusable US-GAAP taxonomy graph with filing-specific XBRL evidence and provides typed tools for retrieval, traversal, computation, and rule evaluation. Two specialized auditors investigate constraints and filing evidence, while a senior auditor coordinates disagreement-triggered refinement. Deterministic tools produce verification evidence; evidential aggregation combines structured reports, including agent-declared overrides, into final decisions. On a newly collected 1{,}000-instance evaluation set of engine-recorded DQC violations, \textsc{AuditFlow} achieves $87.70\%$ joint audit accuracy with Claude Sonnet 4.6, compared with $68.60\%$ for Single Agent and $86.70\%$ for Checker-only. Component and coverage analyses identify executable verification as the main source of accuracy, while additional gains from agent judgment depend on the backbone and checker coverage.

Subjects:	Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA); Symbolic Computation (cs.SC)
Cite as:	arXiv:2606.03031 [cs.AI]
	(or arXiv:2606.03031v2 [cs.AI] for this version)
	https://doi.org/10.48550/arXiv.2606.03031 arXiv-issued DOI via DataCite

Submission history

From: Yan Wang [view email]
[v1] Tue, 2 Jun 2026 02:14:42 UTC (9,452 KB)
[v2] Wed, 14 Oct 2026 01:04:35 UTC (7,795 KB)