MultiFinRAG：一种优化的多模态检索增强生成（RAG）框架，用于金融问答
MultiFinRAG: An Optimized Multimodal Retrieval-Augmented Generation (RAG) Framework for Financial Question Answering
AI总结：
MultiFinRAG通过多模态提取和分层回退策略，提升金融问答任务中跨模态推理的准确性，比ChatGPT-4o在复杂任务上高19个百分点。
摘要：
金融文档，如10-Ks、10-Qs和投资者演示，涵盖数百页内容，包含密集叙述文本、结构化表格和复杂图表。回答此类内容的问题通常需要跨模态联合推理，这对传统大语言模型（LLMs）和检索增强生成（RAG）管道构成挑战，因为存在令牌限制、布局丢失和碎片化跨模态上下文。我们引入MultiFinRAG，一种专为金融问答设计的检索增强生成框架。MultiFinRAG首先通过将表格和图表图像分组并发送到轻量级、量化开源多模态LLM进行多模态提取，生成结构化JSON输出和简洁文本摘要。这些输出与叙述文本一起，通过模态感知相似性阈值进行嵌入和索引以实现精确检索。分层回退策略随后在必要时动态从纯文本升级到文本+表格+图像上下文，实现跨模态推理并减少无关上下文。尽管在通用硬件上运行，MultiFinRAG在涉及文本、表格、图像和混合多模态推理的复杂金融问答任务中，比ChatGPT-4o（免费版）高19个百分点。
英文摘要：
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.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
