Leveraging Expert Guided Adversarial Augmentation For Improving Generalization in Named Entity Recognition

TL;DR

This paper introduces expert-guided adversarial augmentation for NER, significantly enhancing model robustness and out-of-domain generalization by creating challenging training examples and employing mixup regularization.

Contribution

It proposes a novel expert-guided adversarial augmentation method for NER, improving model robustness and out-of-domain performance beyond existing techniques.

Findings

Performance drops significantly on adversarially augmented test set.

Training with augmented data improves robustness and out-of-domain accuracy.

Mixup regularization further enhances generalization.

Abstract

Named Entity Recognition (NER) systems often demonstrate great performance on in-distribution data, but perform poorly on examples drawn from a shifted distribution. One way to evaluate the generalization ability of NER models is to use adversarial examples, on which the specific variations associated with named entities are rarely considered. To this end, we propose leveraging expert-guided heuristics to change the entity tokens and their surrounding contexts thereby altering their entity types as adversarial attacks. Using expert-guided heuristics, we augmented the CoNLL 2003 test set and manually annotated it to construct a high-quality challenging set. We found that state-of-the-art NER systems trained on CoNLL 2003 training data drop performance dramatically on our challenging set. By training on adversarial augmented training examples and using mixup for regularization, we were able to significantly improve the performance on the challenging set as well as improve out-of-domain generalization which we evaluated by using OntoNotes data. We have publicly released our dataset and code at https://github.com/GT-SALT/Guided-Adversarial-Augmentation.