dots.ocr: Multilingual Document Layout Parsing in a Single Vision-Language Model

TL;DR

dots.ocr is a unified vision-language model that jointly learns document layout detection, text recognition, and relational understanding, achieving state-of-the-art results across diverse languages and domains.

Contribution

This paper introduces dots.ocr, the first end-to-end model for multilingual document layout parsing that leverages joint training and a large synthetic dataset.

Findings

Achieves state-of-the-art performance on OmniDocBench.

Demonstrates strong multilingual capabilities across 126 languages.

Improves performance by approximately 10% relative on XDocParse benchmark.

Abstract

Document Layout Parsing serves as a critical gateway for Artificial Intelligence (AI) to access and interpret the world's vast stores of structured knowledge. This process,which encompasses layout detection, text recognition, and relational understanding, is particularly crucial for empowering next-generation Vision-Language Models. Current methods, however, rely on fragmented, multi-stage pipelines that suffer from error propagation and fail to leverage the synergies of joint training. In this paper, we introduce dots_ocr, a single Vision-Language Model that, for the first time, demonstrates the advantages of jointly learning three core tasks within a unified, end-to-end framework. This is made possible by a highly scalable data engine that synthesizes a vast multilingual corpus, empowering the model to deliver robust performance across a wide array of tasks, encompassing diverse languages, layouts, and domains. The efficacy of our unified paradigm is validated by state-of-the-art performance on the comprehensive OmniDocBench. Furthermore, to catalyze research in global document intelligence, we introduce XDocParse, a challenging new benchmark spanning 126 languages. On this benchmark, dots_ocr achieves state-of-the-art performance, delivering an approximately 10% relative improvement and demonstrating strong multilingual capability.