Domain Adversarial Fine-Tuning as an Effective Regularizer

TL;DR

This paper introduces AFTER, a regularization method using domain adversarial training during fine-tuning of language models, which improves performance by preventing overfitting to specific domains.

Contribution

The paper proposes a novel adversarial regularization technique, AFTER, that enhances fine-tuning of pretrained language models by maintaining domain-invariant representations.

Findings

Improved NLP task performance with AFTER compared to standard fine-tuning.

Effective in preventing overfitting to task-specific domains.

Enhances generalization across different domains.

Abstract

In Natural Language Processing (NLP), pretrained language models (LMs) that are transferred to downstream tasks have been recently shown to achieve state-of-the-art results. However, standard fine-tuning can degrade the general-domain representations captured during pretraining. To address this issue, we introduce a new regularization technique, AFTER; domain Adversarial Fine-Tuning as an Effective Regularizer. Specifically, we complement the task-specific loss used during fine-tuning with an adversarial objective. This additional loss term is related to an adversarial classifier, that aims to discriminate between in-domain and out-of-domain text representations. In-domain refers to the labeled dataset of the task at hand while out-of-domain refers to unlabeled data from a different domain. Intuitively, the adversarial classifier acts as a regularizer which prevents the model from overfitting to the task-specific domain. Empirical results on various natural language understanding tasks show that AFTER leads to improved performance compared to standard fine-tuning.