AUDITFLOW: Executable Symbolic Environments for Structured Financial Reporting Verification
The article discusses AuditFlow, a new framework designed for structured financial reporting verification. It utilizes a graph-grounded multi-agent approach to enhance the accuracy of audits by linking reported facts to taxonomy concepts. The framework has demonstrated significant improvements in audit accuracy compared to existing methods.
- ▪AuditFlow separates adaptive search from deterministic verification to improve structured financial audit processes.
- ▪The framework builds a symbolic environment using a static US-GAAP taxonomy graph and a dynamic XBRL filing graph.
- ▪AuditFlow achieved 82.09% joint audit accuracy under GPT-5.5, outperforming the strongest baseline by 14.93 points.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Artificial Intelligence arXiv:2606.03031 (cs) [Submitted on 2 Jun 2026] Title:AUDITFLOW: Executable Symbolic Environments for Structured Financial Reporting Verification Authors:Yan Wang, Xuguang Ai, Jaisal Patel, Xueqing Peng, Fengran Mo, Yupeng Cao, Haohang Li, Mingyu Cao, Lingfei Qian, Víctor Gutiérrez-Basulto View a PDF of the paper titled AUDITFLOW: Executable Symbolic Environments for Structured Financial Reporting Verification, by Yan Wang and 9 other authors View PDF HTML (experimental) Abstract:Structured financial audit verification is difficult for language-model agents because correctness depends on structured evidence rather than text alone.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
