The field of document understanding is moving towards a more holistic approach, incorporating multimodal retrieval and reasoning to unlock comprehensive document intelligence. This shift is driven by the need to overcome the limitations of current approaches, which either lose structural detail or struggle with context modeling. The use of Retrieval-Augmented Generation (RAG) is becoming increasingly prominent, enabling models to ground themselves in external data and improve their performance. Noteworthy papers in this area include: Scaling Beyond Context, which presents a systematic survey of Multimodal RAG for document understanding. Fine-Tuning MedGemma for Clinical Captioning, which proposes a framework to specialize the MedGemma model for generating high-fidelity captions. Towards Mixed-Modal Retrieval for Universal Retrieval-Augmented Generation, which introduces a unified mixed-modal to mixed-modal retriever tailored for Universal Retrieval-Augmented Generation (URAG) scenarios.
Multimodal Retrieval-Augmented Generation for Document Understanding
Sources
Scaling Beyond Context: A Survey of Multimodal Retrieval-Augmented Generation for Document Understanding
Fine-Tuning MedGemma for Clinical Captioning to Enhance Multimodal RAG over Malaysia CPGs
Demo: Guide-RAG: Evidence-Driven Corpus Curation for Retrieval-Augmented Generation in Long COVID
EMRRG: Efficient Fine-Tuning Pre-trained X-ray Mamba Networks for Radiology Report Generation
Towards Mixed-Modal Retrieval for Universal Retrieval-Augmented Generation
Multimedia-Aware Question Answering: A Review of Retrieval and Cross-Modal Reasoning Architectures
VLSP 2025 MLQA-TSR Challenge: Vietnamese Multimodal Legal Question Answering on Traffic Sign Regulation