CounterGeDi: A controllable approach to generate polite, detoxified and emotional counterspeech

TL;DR

CounterGeDi is a controllable generation model that guides counterspeech production to be more polite, detoxified, and emotionally expressive, improving quality without losing relevance across multiple datasets.

Contribution

It introduces CounterGeDi, an ensemble of generative discriminators, to steer counterspeech generation towards specific attributes, enhancing effectiveness over vanilla models.

Findings

Politeness scores increased by ~15%.

Detoxification scores increased by ~6%.

Emotion in counterspeech increased by at least 10%.

Abstract

Recently, many studies have tried to create generation models to assist counter speakers by providing counterspeech suggestions for combating the explosive proliferation of online hate. However, since these suggestions are from a vanilla generation model, they might not include the appropriate properties required to counter a particular hate speech instance. In this paper, we propose CounterGeDi - an ensemble of generative discriminators (GeDi) to guide the generation of a DialoGPT model toward more polite, detoxified, and emotionally laden counterspeech. We generate counterspeech using three datasets and observe significant improvement across different attribute scores. The politeness and detoxification scores increased by around 15% and 6% respectively, while the emotion in the counterspeech increased by at least 10% across all the datasets. We also experiment with triple-attribute control and observe significant improvement over single attribute results when combining complementing attributes, e.g., politeness, joyfulness and detoxification. In all these experiments, the relevancy of the generated text does not deteriorate due to the application of these controls