LayoutLM is a simple but effective pre-training method of text and layout for document image understanding and information extraction tasks, such as form understanding and receipt understanding. LayoutLM archives the SOTA results on multiple datasets. For more details, please refer to our paper:
We pre-train LayoutLM on IIT-CDIP Test Collection 1.0* dataset with two settings.
1@misc{xu2019layoutlm,
2 title={LayoutLM: Pre-training of Text and Layout for Document Image Understanding},
3 author={Yiheng Xu and Minghao Li and Lei Cui and Shaohan Huang and Furu Wei and Ming Zhou},
4 year={2019},
5 eprint={1912.13318},
6 archivePrefix={arXiv},
7 primaryClass={cs.CL}
8}
