发表机构
BMO Financial Group（蒙特利尔银行金融集团）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对财务文档验证的异构性与业务规则处理难题，提出基于多模态大语言模型的LAVA框架，通过四阶段设计提升幻觉控制与边缘案例处理能力，适用于高风险财务审计场景。
AI 中文摘要
生产环境中的财务文档验证（如工资单审计、税务合规、贷款承保）需在严格的企业约束下具备极高的准确性、一致性和可复现性。实际应用中，文档存在异构布局与格式、语义丰富且依赖上下文的内容，以及嵌入的业务规则，现有流程难以可靠处理。我们提出LAVA（Logic-Aware Validation and Augmentation，逻辑感知验证与增强），这是一个基于多模态大语言模型的模块化、与主干无关的流程，包含四个阶段设计：文档-规则检索、布局保留的信息提取、辅助元数据丰富，以及可审计的符号/算术验证。LAVA支持稳健的规则 grounding、细粒度错误归因，以及一致、可追溯的端到端执行，这些是高风险部署的关键能力。在包含多样财务文档和数十条专家整理验证规则的大型真实基准上评估，LAVA在幻觉控制和边缘案例处理方面优于基线，同时保持高效的token使用，证明了其在高容量、时间关键型验证中的实用性。
英文摘要
Financial document validation in production, such as payroll auditing, tax compliance, and loan underwriting, demands exceptional accuracy, consistency, and reproducibility under strict enterprise constraints. In practice, documents arrive with heterogeneous layouts and formats, semantically rich and context-dependent content, and embedded business rules that current pipelines struggle to process reliably. We introduce LAVA (Logic-Aware Validation and Augmentation), a modular, backbone-agnostic pipeline built on multimodal large language models, that integrates a four-stage design: document-rule retrieval, layout-preserving information extraction, auxiliary metadata enrichment, and auditable symbolic/arithmetic verification. LAVA supports robust rule grounding, fine-grained error attribution, and consistent, traceable end-to-end execution, capabilities essential for high-stakes deployment. Evaluated on a large real-world benchmark with diverse financial documents and dozens of expert-curated validation rules, LAVA outperforms baselines in hallucination control and edge-case handling while maintaining efficient token usage, demonstrating practicality for high-volume, time-critical validation.
Journal refProceedings of The 10th Workshop on Financial Technology and Natural Language Processing (FinNLP 2025), Association for Computational Linguistics, pp. 75-92, 2025
DOI：10.18653/v1/2025.finnlp-2.7
