Responsibility Perspective Transfer for Italian Femicide News

TL;DR

This paper introduces a novel task of automatically rewriting gender-based violence descriptions to alter perceived responsibility, aiming to influence perceptions and reduce stereotypes, using a new dataset and multiple NLP methods evaluated through human and automatic metrics.

Contribution

It presents the first dataset and approach for perspective-based rewriting of GBV descriptions to modify perceived responsibility levels.

Findings

Unsupervised, zero-shot, and few-shot models can effectively rewrite sentences.

Models can alter perceived responsibility as shown by human evaluations.

Automatic metrics correlate with human judgments of rewriting quality.

Abstract

Different ways of linguistically expressing the same real-world event can lead to different perceptions of what happened. Previous work has shown that different descriptions of gender-based violence (GBV) influence the reader's perception of who is to blame for the violence, possibly reinforcing stereotypes which see the victim as partly responsible, too. As a contribution to raise awareness on perspective-based writing, and to facilitate access to alternative perspectives, we introduce the novel task of automatically rewriting GBV descriptions as a means to alter the perceived level of responsibility on the perpetrator. We present a quasi-parallel dataset of sentences with low and high perceived responsibility levels for the perpetrator, and experiment with unsupervised (mBART-based), zero-shot and few-shot (GPT3-based) methods for rewriting sentences. We evaluate our models using a questionnaire study and a suite of automatic metrics.