Community Norms in the Spotlight: Enabling Task-Agnostic Unsupervised Pre-Training to Benefit Online Social Media

TL;DR

This paper proposes a novel unsupervised pretraining framework based on community norms to improve modeling of online social media dynamics, reducing reliance on labeled data and enhancing interpretability.

Contribution

It introduces a community norms-based pretraining paradigm that shifts from task-specific fine-tuning to a more general, interpretable approach for social media modeling.

Findings

Reduces dependence on labeled datasets.

Enhances interpretability of social norms.

Potential for improved social media moderation.

Abstract

Modelling the complex dynamics of online social platforms is critical for addressing challenges such as hate speech and misinformation. While Discussion Transformers, which model conversations as graph structures, have emerged as a promising architecture, their potential is severely constrained by reliance on high-quality, human-labelled datasets. In this paper, we advocate a paradigm shift from task-specific fine-tuning to unsupervised pretraining, grounded in an entirely novel consideration of community norms. We posit that this framework not only mitigates data scarcity but also enables interpretation of the social norms underlying the decisions made by such an AI system. Ultimately, we believe that this direction offers many opportunities for AI for Social Good.