# PP-DocBee2: Improved Baselines with Efficient Data for Multimodal Document Understanding

arXiv: https://arxiv.org/abs/2506.18023 · HTML: https://mlsanity.com/paper/2506.18023/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 5 of its 7 references that are in the map.
- First posted on arXiv 22 Jun 2025.

## Builds on (most cited first, 5 of 5)

- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling](https://mlsanity.com/paper/2412.05271/) (2024)
- [ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools](https://mlsanity.com/paper/2406.12793/) (2024)
- [Multimodal Language Models See Better When They Look Shallower](https://mlsanity.com/paper/2504.21447/) (2025)
- [PP-DocBee: Improving Multimodal Document Understanding Through a Bag of Tricks](https://mlsanity.com/paper/2503.04065/) (2025)

## Abstract (arXiv)

This report introduces PP-DocBee2, an advanced version of the PP-DocBee, designed to enhance multimodal document understanding. Built on a large multimodal model architecture, PP-DocBee2 addresses the limitations of its predecessor through key technological improvements, including enhanced synthetic data quality, improved visual feature fusion strategy, and optimized inference methodologies. These enhancements yield an 11.4% performance boost on internal benchmarks for Chinese business documents, and reduce inference latency by 73.0% to the vanilla version. A key innovation of our work is a data quality optimization strategy for multimodal document tasks. By employing a large-scale multimodal pre-trained model to evaluate data, we apply a novel statistical criterion to filter outliers, ensuring high-quality training data. Inspired by insights into underutilized intermediate features in multimodal models, we enhance the ViT representational capacity by decomposing it into layers and applying a novel feature fusion strategy to improve complex reasoning. The source code and pre-trained model are available at hrefhttps://github.com/PaddlePaddle/PaddleMIXhttps://github.com/PaddlePaddle/PaddleMIX.