We propose a novel framework to conduct field extraction from forms with unlabeled data. To bootstrap the training process, we develop a rule-based method for mining noisy pseudo-labels from unlabeled forms. Using the supervisory signal from the pseudo-labels, we extract a discriminative token representation from a transformer-based model by modeling the interaction between text in the form. To prevent the model from overfitting to label noise, we introduce a refinement module based on a progressive pseudo-label ensemble. Experimental results demonstrate the effectiveness of our framework.
@article{arxiv.2110.04282,
title = {Field Extraction from Forms with Unlabeled Data},
author = {Mingfei Gao and Zeyuan Chen and Nikhil Naik and Kazuma Hashimoto and Caiming Xiong and Ran Xu},
journal= {arXiv preprint arXiv:2110.04282},
year = {2022}
}