# BigDocs: An Open Dataset for Training Multimodal Models on Document and Code Tasks

arXiv: https://arxiv.org/abs/2412.04626 · HTML: https://mlsanity.com/paper/2412.04626/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Dec 2024.

## Built on by (most cited first, 12 of 12)

- [Paper2Poster: Towards Multimodal Poster Automation from Scientific Papers](https://mlsanity.com/paper/2505.21497/) (2025)
- [Qianfan-OCR: A Unified End-to-End Model for Document Intelligence](https://mlsanity.com/paper/2603.13398/) (2026)
- [Breaking the SFT Plateau: Multimodal Structured Reinforcement Learning for Chart-to-Code Generation](https://mlsanity.com/paper/2508.13587/) (2025)
- [BigCharts-R1: Enhanced Chart Reasoning with Visual Reinforcement Finetuning](https://mlsanity.com/paper/2508.09804/) (2025)
- [Qianfan-VL: Domain-Enhanced Universal Vision-Language Models](https://mlsanity.com/paper/2509.18189/) (2025)
- [Beyond NL2Code: A Structured Survey of Multimodal Code Intelligence](https://mlsanity.com/paper/2606.15932/) (2026)
- [VectorGym: A Multi-Task Benchmark for SVG Code Generation, Sketching and Editing](https://mlsanity.com/paper/2603.29852/) (2026)
- [Any2Poster: Any-Source Poster Generation Across Modalities and Domains](https://mlsanity.com/paper/2606.02915/) (2026)
- [Scope: Selective Cross-modal Orchestration of Visual Perception Experts](https://mlsanity.com/paper/2510.12974/) (2025)
- [FORGE: Fine-grained Multimodal Evaluation for Manufacturing Scenarios](https://mlsanity.com/paper/2604.07413/) (2026)
- [Seeing Straight: Document Orientation Detection for Efficient OCR](https://mlsanity.com/paper/2511.04161/) (2025)
- [Question-Guided Evidence Acquisition for Multimodal Visual Question Answering](https://mlsanity.com/paper/2608.19739/) (2026)

## Abstract (arXiv)

Multimodal AI has the potential to significantly enhance document-understanding tasks, such as processing receipts, understanding workflows, extracting data from documents, and summarizing reports. Code generation tasks that require long-structured outputs can also be enhanced by multimodality. Despite this, their use in commercial applications is often limited due to limited access to training data and restrictive licensing, which hinders open access. To address these limitations, we introduce BigDocs-7.5M, a high-quality, open-access dataset comprising 7.5 million multimodal documents across 30 tasks. We use an efficient data curation process to ensure our data is high-quality and license-permissive. Our process emphasizes accountability, responsibility, and transparency through filtering rules, traceable metadata, and careful content analysis. Additionally, we introduce BigDocs-Bench, a benchmark suite with 10 novel tasks where we create datasets that reflect real-world use cases involving reasoning over Graphical User Interfaces (GUI) and code generation from images. Our experiments show that training with BigDocs-Bench improves average performance up to 25.8% over closed-source GPT-4o in document reasoning and structured output tasks such as Screenshot2HTML or Image2Latex generation. Finally, human evaluations showed a preference for outputs from models trained on BigDocs over GPT-4o. This suggests that BigDocs can help both academics and the open-source community utilize and improve AI tools to enhance multimodal capabilities and document reasoning. The project is hosted at https://bigdocs.github.io .