ReMask: A Robust Information-Masking Approach for Domain Counterfactual Generation

by   Pengfei Hong, et al.
IIT Kharagpur
Singapore University of Technology and Design

Domain shift is a big challenge in NLP, thus, many approaches resort to learning domain-invariant features to mitigate the inference phase domain shift. Such methods, however, fail to leverage the domain-specific nuances relevant to the task at hand. To avoid such drawbacks, domain counterfactual generation aims to transform a text from the source domain to a given target domain. However, due to the limited availability of data, such frequency-based methods often miss and lead to some valid and spurious domain-token associations. Hence, we employ a three-step domain obfuscation approach that involves frequency and attention norm-based masking, to mask domain-specific cues, and unmasking to regain the domain generic context. Our experiments empirically show that the counterfactual samples sourced from our masked text lead to improved domain transfer on 10 out of 12 domain sentiment classification settings, with an average of 2 state-of-the-art for unsupervised domain adaptation (UDA). Further, our model outperforms the state-of-the-art by achieving 1.4 in the adversarial domain adaptation (ADA) setting. Moreover, our model also shows its domain adaptation efficacy on a large multi-domain intent classification dataset where it attains state-of-the-art results. We release the codes publicly at <>.


page 1

page 2

page 3

page 4


UDApter – Efficient Domain Adaptation Using Adapters

We propose two methods to make unsupervised domain adaptation (UDA) more...

Interventional Domain Adaptation

Domain adaptation (DA) aims to transfer discriminative features learned ...

Light-weight Calibrator: a Separable Component for Unsupervised Domain Adaptation

Existing domain adaptation methods aim at learning features that can be ...

Unsupervised Domain Adaptation: An Adaptive Feature Norm Approach

Unsupervised domain adaptation aims to mitigate the domain shift when tr...

Modular Domain Adaptation

Off-the-shelf models are widely used by computational social science res...

Counterfactually Guided Policy Transfer in Clinical Settings

Reliably transferring treatment policies learned in one clinical environ...

DoCoGen: Domain Counterfactual Generation for Low Resource Domain Adaptation

Natural language processing (NLP) algorithms have become very successful...

Please sign up or login with your details

Forgot password? Click here to reset