Towards Robust and Privacy-preserving Text Representations

TL;DR

This paper introduces a method to make text representations invariant to author attributes, enhancing privacy and robustness across different datasets and conditions.

Contribution

It proposes a training-time approach to obscure author characteristics, improving privacy and model robustness in text representations.

Findings

Increased privacy of learned representations.

Enhanced robustness to out-of-domain data.

Invariant representations to author attributes.

Abstract

Written text often provides sufficient clues to identify the author, their gender, age, and other important attributes. Consequently, the authorship of training and evaluation corpora can have unforeseen impacts, including differing model performance for different user groups, as well as privacy implications. In this paper, we propose an approach to explicitly obscure important author characteristics at training time, such that representations learned are invariant to these attributes. Evaluating on two tasks, we show that this leads to increased privacy in the learned representations, as well as more robust models to varying evaluation conditions, including out-of-domain corpora.