AgenticOCR: Parsing Only What You Need for Efficient Retrieval-Augmented Generation

TL;DR

AgenticOCR introduces a query-driven, selective OCR parsing method that enhances the efficiency and accuracy of retrieval-augmented generation for complex visual documents by focusing only on relevant regions.

Contribution

This paper presents a novel dynamic OCR parsing paradigm that enables on-demand, region-specific recognition, improving long document understanding in visual RAG systems.

Findings

Improves efficiency of visual RAG systems

Achieves expert-level performance in long document understanding

Reduces extraneous context and hallucinations

Abstract

The expansion of retrieval-augmented generation (RAG) into multimodal domains has intensified the challenge for processing complex visual documents, such as financial reports. While page-level chunking and retrieval is a natural starting point, it creates a critical bottleneck: delivering entire pages to the generator introduces excessive extraneous context. This not only overloads the generator's attention mechanism but also dilutes the most salient evidence. Moreover, compressing these information-rich pages into a limited visual token budget further increases the risk of hallucinations. To address this, we introduce AgenticOCR, a dynamic parsing paradigm that transforms optical character recognition (OCR) from a static, full-text process into a query-driven, on-demand extraction system. By autonomously analyzing document layout in a "thinking with images" manner, AgenticOCR identifies and selectively recognizes regions of interest. This approach performs on-demand decompression of visual tokens precisely where needed, effectively decoupling retrieval granularity from rigid page-level chunking. AgenticOCR has the potential to serve as the "third building block" of the visual document RAG stack, operating alongside and enhancing standard Embedding and Reranking modules. Experimental results demonstrate that AgenticOCR improves both the efficiency and accuracy of visual RAG systems, achieving expert-level performance in long document understanding. Code and models are available at https://github.com/OpenDataLab/AgenticOCR.