Evidence-Driven Retrieval Augmented Response Generation for Online Misinformation

TL;DR

This paper introduces RARG, a retrieval-augmented method that collects scientific evidence and uses large language models to generate polite, factual responses refuting online misinformation, especially demonstrated on COVID-19 data.

Contribution

The paper presents a novel two-stage framework combining evidence retrieval and reinforcement learning-based response generation for countering misinformation.

Findings

RARG outperforms baseline methods in response quality.

The approach effectively refutes misinformation with evidence-based responses.

It maintains politeness and factual accuracy in generated responses.

Abstract

The proliferation of online misinformation has posed significant threats to public interest. While numerous online users actively participate in the combat against misinformation, many of such responses can be characterized by the lack of politeness and supporting facts. As a solution, text generation approaches are proposed to automatically produce counter-misinformation responses. Nevertheless, existing methods are often trained end-to-end without leveraging external knowledge, resulting in subpar text quality and excessively repetitive responses. In this paper, we propose retrieval augmented response generation for online misinformation (RARG), which collects supporting evidence from scientific sources and generates counter-misinformation responses based on the evidences. In particular, our RARG consists of two stages: (1) evidence collection, where we design a retrieval pipeline to retrieve and rerank evidence documents using a database comprising over 1M academic articles; (2) response generation, in which we align large language models (LLMs) to generate evidence-based responses via reinforcement learning from human feedback (RLHF). We propose a reward function to maximize the utilization of the retrieved evidence while maintaining the quality of the generated text, which yields polite and factual responses that clearly refutes misinformation. To demonstrate the effectiveness of our method, we study the case of COVID-19 and perform extensive experiments with both in- and cross-domain datasets, where RARG consistently outperforms baselines by generating high-quality counter-misinformation responses.