Improving Authorship Verification using Linguistic Divergence

TL;DR

This paper introduces an unsupervised method for authorship verification using a novel DV-Distance metric derived from pre-trained deep language models, effectively addressing non-comparability issues especially in small or cross-domain datasets.

Contribution

It presents the first method explicitly designed to handle non-comparability in authorship verification, leveraging deep language models for improved accuracy.

Findings

Matches or surpasses state-of-the-art performance on four datasets

Addresses non-comparability in small or cross-domain corpora

Provides an intuitive, visualizable approach

Abstract

We propose an unsupervised solution to the Authorship Verification task that utilizes pre-trained deep language models to compute a new metric called DV-Distance. The proposed metric is a measure of the difference between the two authors comparing against pre-trained language models. Our design addresses the problem of non-comparability in authorship verification, frequently encountered in small or cross-domain corpora. To the best of our knowledge, this paper is the first one to introduce a method designed with non-comparability in mind from the ground up, rather than indirectly. It is also one of the first to use Deep Language Models in this setting. The approach is intuitive, and it is easy to understand and interpret through visualization. Experiments on four datasets show our methods matching or surpassing current state-of-the-art and strong baselines in most tasks.