This directory contains the data for the paper PDF-WuKong: A Large Multimodal Model for Efficient Long PDF Reading with End-to-End Sparse Sampling
and a code for extracting text and image information from XML documents:
The structure of this repository is shown as follows.
PDF-Wukong
β
β
βββ PaperPDF.pyβ¦ See the full description on the dataset page:
https://huggingface.co/datasets/yh0075/PaperPDF.