Effective Unsupervised Domain Adaptation with Adversarially Trained Language Models

TL;DR

This paper introduces an adversarial masking strategy for unsupervised domain adaptation of language models, significantly improving performance by focusing training on domain-specific tokens.

Contribution

It proposes a novel adversarial masking approach combined with a dynamic programming optimization to enhance domain adaptation of masked language models.

Findings

Achieves up to +1.64 F1 score improvement on NER tasks.

Outperforms random masking strategies in domain adaptation.

Effective in six different unsupervised domain adaptation scenarios.

Abstract

Recent work has shown the importance of adaptation of broad-coverage contextualised embedding models on the domain of the target task of interest. Current self-supervised adaptation methods are simplistic, as the training signal comes from a small percentage of \emph{randomly} masked-out tokens. In this paper, we show that careful masking strategies can bridge the knowledge gap of masked language models (MLMs) about the domains more effectively by allocating self-supervision where it is needed. Furthermore, we propose an effective training strategy by adversarially masking out those tokens which are harder to reconstruct by the underlying MLM. The adversarial objective leads to a challenging combinatorial optimisation problem over \emph{subsets} of tokens, which we tackle efficiently through relaxation to a variational lowerbound and dynamic programming. On six unsupervised domain adaptation tasks involving named entity recognition, our method strongly outperforms the random masking strategy and achieves up to +1.64 F1 score improvements.