Unveil: Unified Visual-Textual Integration and Distillation for Multi-modal Document Retrieval

2026-05-23

AI Summary

This work addresses the challenge of document retrieval in real-world scenarios where formats are highly diverse. Traditional text-based methods often disregard layout information, while purely visual models struggle to capture fine-grained semantic content. To bridge this gap, the authors propose Unveil, a novel framework that jointly optimizes vision–text embeddings and knowledge distillation to efficiently transfer multimodal semantic knowledge into a pure vision model, enabling high-performance retrieval without explicit document parsing. By integrating multimodal alignment with parse-free representation learning, Unveil preserves semantic fidelity while significantly improving both retrieval accuracy and efficiency, effectively narrowing the performance gap between multimodal and purely visual approaches.

Abstract

Document retrieval in real-world scenarios faces significant challenges due to diverse document formats and modalities. Traditional text-based approaches rely on tailored parsing techniques that disregard layout information and are prone to errors, while recent parsing-free visual methods often struggle to capture fine-grained textual semantics in text-rich scenarios. To address these limitations, we propose \textbf{Unveil}, a novel visual-textual embedding framework that effectively integrates textual and visual features for robust document representation. Through knowledge distillation, we transfer the semantic understanding capabilities from the visual-textual embedding model to a purely visual model, enabling efficient parsing-free retrieval while preserving semantic fidelity. Experimental results demonstrate that our visual-textual embedding method surpasses existing approaches, while knowledge distillation successfully bridges the performance gap between visual-textual and visual-only methods, improving both retrieval accuracy and efficiency.

Problem

• document retrieval
• multi-modal
• visual-textual integration
• parsing-free
• text-rich documents

Innovation

• visual-textual integration
• knowledge distillation
• parsing-free retrieval
• multi-modal document retrieval
• document representation