Can Authorship Representation Learning Capture Stylistic Features?

TL;DR

This paper investigates whether authorship representations learned from large text corpora primarily encode writing style, using targeted experiments to validate their sensitivity to stylistic features and potential robustness to topic shifts.

Contribution

The study systematically probes authorship representations to confirm they mainly encode stylistic features, informing future stylistic applications and robustness considerations.

Findings

Authorship representations are sensitive to writing style.

Representations may be robust to topic drift over time.

Potential for stylistic applications like style transfer.

Abstract

Automatically disentangling an author's style from the content of their writing is a longstanding and possibly insurmountable problem in computational linguistics. At the same time, the availability of large text corpora furnished with author labels has recently enabled learning authorship representations in a purely data-driven manner for authorship attribution, a task that ostensibly depends to a greater extent on encoding writing style than encoding content. However, success on this surrogate task does not ensure that such representations capture writing style since authorship could also be correlated with other latent variables, such as topic. In an effort to better understand the nature of the information these representations convey, and specifically to validate the hypothesis that they chiefly encode writing style, we systematically probe these representations through a series of targeted experiments. The results of these experiments suggest that representations learned for the surrogate authorship prediction task are indeed sensitive to writing style. As a consequence, authorship representations may be expected to be robust to certain kinds of data shift, such as topic drift over time. Additionally, our findings may open the door to downstream applications that require stylistic representations, such as style transfer.