EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

HierFinRAG—Hierarchical Multimodal RAG for Financial Document Understanding

Quang-Vinh Dang, Ngoc-Son-An Nguyen, Thi-Bich-Diem Vo

Informatics · 10.02.2026 · 13 (2) · 30

DOI: 10.3390/informatics13020030 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Informatics, online erschienen am 10.02.2026. Verfasst von Quang-Vinh Dang und 2 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 65 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Financial document understanding remains a critical challenge for Large Language Models, primarily due to the complex interplay between narrative text and structured numerical tables. Existing Retrieval-Augmented Generation (RAG) systems often treat these modalities in isolation, leading to significant failures in tasks requiring joint reasoning. This study introduces HierFinRAG, a novel hierarchical multimodal framework designed to unify tabular and textual data processing. Our approach employs a Table-Text Graph Neural Network (TTGNN) to explicitly model semantic and structural dependencies between table cells and corresponding text, coupled with a Symbolic–Neural Fusion module that routes queries between a neural generator and a symbolic calculator for precise arithmetic operations. We evaluate the system on the FinQA and FinanceBench datasets, comparing performance against strong baselines including Vanilla RAG and GPT-4o with Code Interpreter. Results demonstrate that HierFinRAG achieves an Exact Match score of 82.5% on FinQA, surpassing the best baseline by 6.5 percentage points, while maintaining a 3.5× faster inference latency than agentic approaches. These findings indicate that integrating hierarchical structural awareness with hybrid reasoning significantly enhances the accuracy and interpretability of financial artificial intelligence systems.

Zitieren

APA

Dang, Q., Nguyen, N., & Vo, T. (2026). HierFinRAG—Hierarchical Multimodal RAG for Financial Document Understanding. Informatics, 13(2), 30. https://doi.org/10.3390/informatics13020030

Vancouver

Dang Q, Nguyen N, Vo T. HierFinRAG—Hierarchical Multimodal RAG for Financial Document Understanding. Informatics. 2026;13(2):30. doi:10.3390/informatics13020030

BibTeX
@article{Dang2026,
  author = {Dang, Quang-Vinh and Nguyen, Ngoc-Son-An and Vo, Thi-Bich-Diem},
  title = {HierFinRAG—Hierarchical Multimodal RAG for Financial Document Understanding},
  journal = {Informatics},
  year = {2026},
  volume = {13},
  number = {2},
  pages = {30},
  doi = {10.3390/informatics13020030}
}
Kostenlos starten →