# Agentic Retrieval-Augmented Generation for Financial Document Question Answering Source: [Agentic Retrieval-Augmented Generation for Financial Document Question Answering]() Collection index: [llms.txt]() Data updated: 2026-07-08T05:00:20.838Z Authors: Yang Shu, Yingmin Liu, Zequn Xie Published: 2026-05-06T19:59:51.000Z ## Abstract Financial document question answering (QA) demands complex multi-step numerical reasoning over heterogeneous evidence--structured tables, textual narratives, and footnotes--scattered across corporate filings. Existing retrieval-augmented generation (RAG) approaches adopt a single-pass retrieve-then-generate paradigm that struggles with the compositional reasoning chains prevalent in financial analysis. We propose FinAgent-RAG, an agentic RAG framework that orchestrates iterative retrieval-reasoning loops with self-verification, specifically engineered for the precision requirements of financial numerical reasoning. The framework integrates three domain-specific innovations: (1) a Contrastive Financial Retriever trained with hard negative mining to distinguish semantically similar but numerically distinct financial passages, (2) a Program-of-Thought reasoning module that generates executable Python code for precise arithmetic rather than relying on error-prone LLM-based mental computation, and (3) an Adaptive Strategy Router that dynamically allocates computational resources based on question complexity, reducing API costs by 41.3% on FinQA while preserving accuracy. Extensive experiments on three benchmark datasets--FinQA, ConvFinQA, and TAT-QA--demonstrate that FinAgent-RAG achieves 76.81%, 78.46%, and 74.96% execution accuracy respectively, outperforming the strongest baseline by 5.62--9.32 percentage points. Ablation studies, cross-backbone evaluation with four LLMs, and deployment cost analysis confirm the framework's robustness and practical viability for financial institutions. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Financial Document QA on FinQA (test)](): Financial Document QA; FinQA (test); Execution Accuracy: 76.81. - [Financial Question Answering on ConvFinQA (test)](): Financial Question Answering; ConvFinQA (test); Execution Accuracy: 78.46. - [Financial Question Answering on TAT-QA (test)](): Financial Question Answering; TAT-QA (test); Execution Accuracy: 74.96. ## Original sources and implementation [Original paper]() [PDF]()