Re-ranking Using Large Language Models for Mitigating Exposure to Harmful Content on Social Media Platforms

TL;DR

This paper introduces a novel LLM-based re-ranking method that reduces harmful content exposure on social media, offering a scalable and adaptable alternative to traditional moderation techniques.

Contribution

It presents a new re-ranking approach using Large Language Models in zero-shot and few-shot settings to mitigate harmful content without extensive labeled data.

Findings

Outperforms existing moderation approaches in experiments

Effectively reduces exposure to harmful content across datasets and models

Introduces new metrics for evaluating harm mitigation effectiveness

Abstract

Social media platforms utilize Machine Learning (ML) and Artificial Intelligence (AI) powered recommendation algorithms to maximize user engagement, which can result in inadvertent exposure to harmful content. Current moderation efforts, reliant on classifiers trained with extensive human-annotated data, struggle with scalability and adapting to new forms of harm. To address these challenges, we propose a novel re-ranking approach using Large Language Models (LLMs) in zero-shot and few-shot settings. Our method dynamically assesses and re-ranks content sequences, effectively mitigating harmful content exposure without requiring extensive labeled data. Alongside traditional ranking metrics, we also introduce two new metrics to evaluate the effectiveness of re-ranking in reducing exposure to harmful content. Through experiments on three datasets, three models and across three configurations, we demonstrate that our LLM-based approach significantly outperforms existing proprietary moderation approaches, offering a scalable and adaptable solution for harm mitigation.