MultiFinRAG: An Optimized Multimodal Retrieval-Augmented Generation (RAG) Framework for Financial Question Answering

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gondhalekar, Chinmay, Patel, Urjitkumar, Yeh, Fang-Chun
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918437135056896
author Gondhalekar, Chinmay
Patel, Urjitkumar
Yeh, Fang-Chun
author_facet Gondhalekar, Chinmay
Patel, Urjitkumar
Yeh, Fang-Chun
contents Financial documents--such as 10-Ks, 10-Qs, and investor presentations--span hundreds of pages and combine diverse modalities, including dense narrative text, structured tables, and complex figures. Answering questions over such content often requires joint reasoning across modalities, which strains traditional large language models (LLMs) and retrieval-augmented generation (RAG) pipelines due to token limitations, layout loss, and fragmented cross-modal context. We introduce MultiFinRAG, a retrieval-augmented generation framework purpose-built for financial QA. MultiFinRAG first performs multimodal extraction by grouping table and figure images into batches and sending them to a lightweight, quantized open-source multimodal LLM, which produces both structured JSON outputs and concise textual summaries. These outputs, along with narrative text, are embedded and indexed with modality-aware similarity thresholds for precise retrieval. A tiered fallback strategy then dynamically escalates from text-only to text+table+image contexts when necessary, enabling cross-modal reasoning while reducing irrelevant context. Despite running on commodity hardware, MultiFinRAG achieves 19 percentage points higher accuracy than ChatGPT-4o (free-tier) on complex financial QA tasks involving text, tables, images, and combined multimodal reasoning.
format Preprint
id arxiv_https___arxiv_org_abs_2506_20821
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MultiFinRAG: An Optimized Multimodal Retrieval-Augmented Generation (RAG) Framework for Financial Question Answering
Gondhalekar, Chinmay
Patel, Urjitkumar
Yeh, Fang-Chun
Computation and Language
Artificial Intelligence
Computational Engineering, Finance, and Science
68T50, 68T07 (Primary) 68P20, 91G15, 91G70, 68U10 (Secondary)
I.2.7; I.2.10; H.3.3; H.2.8; I.5.4; J.1
Financial documents--such as 10-Ks, 10-Qs, and investor presentations--span hundreds of pages and combine diverse modalities, including dense narrative text, structured tables, and complex figures. Answering questions over such content often requires joint reasoning across modalities, which strains traditional large language models (LLMs) and retrieval-augmented generation (RAG) pipelines due to token limitations, layout loss, and fragmented cross-modal context. We introduce MultiFinRAG, a retrieval-augmented generation framework purpose-built for financial QA. MultiFinRAG first performs multimodal extraction by grouping table and figure images into batches and sending them to a lightweight, quantized open-source multimodal LLM, which produces both structured JSON outputs and concise textual summaries. These outputs, along with narrative text, are embedded and indexed with modality-aware similarity thresholds for precise retrieval. A tiered fallback strategy then dynamically escalates from text-only to text+table+image contexts when necessary, enabling cross-modal reasoning while reducing irrelevant context. Despite running on commodity hardware, MultiFinRAG achieves 19 percentage points higher accuracy than ChatGPT-4o (free-tier) on complex financial QA tasks involving text, tables, images, and combined multimodal reasoning.
title MultiFinRAG: An Optimized Multimodal Retrieval-Augmented Generation (RAG) Framework for Financial Question Answering
topic Computation and Language
Artificial Intelligence
Computational Engineering, Finance, and Science
68T50, 68T07 (Primary) 68P20, 91G15, 91G70, 68U10 (Secondary)
I.2.7; I.2.10; H.3.3; H.2.8; I.5.4; J.1
url https://arxiv.org/abs/2506.20821