RAEmoLLM: Retrieval Augmented LLMs for Cross-Domain Misinformation Detection Using In-Context Learning Based on Emotional Information

TL;DR

RAEmoLLM is a retrieval-augmented large language model framework that leverages emotional and affective information for improved cross-domain misinformation detection without the need for fine-tuning.

Contribution

It introduces a novel retrieval-augmented LLM approach that incorporates affective embeddings for cross-domain misinformation detection, eliminating the need for resource-intensive fine-tuning.

Findings

Significant performance improvements on three misinformation benchmarks.

Achieved up to 31.18% accuracy increase over baseline methods.

Effectively utilizes affective embeddings for cross-domain generalization.

Abstract

Misinformation is prevalent in various fields such as education, politics, health, etc., causing significant harm to society. However, current methods for cross-domain misinformation detection rely on effort- and resource-intensive fine-tuning and complex model structures. With the outstanding performance of LLMs, many studies have employed them for misinformation detection. Unfortunately, they focus on in-domain tasks and do not incorporate significant sentiment and emotion features (which we jointly call {\em affect}). In this paper, we propose RAEmoLLM, the first retrieval augmented (RAG) LLMs framework to address cross-domain misinformation detection using in-context learning based on affective information. RAEmoLLM includes three modules. (1) In the index construction module, we apply an emotional LLM to obtain affective embeddings from all domains to construct a retrieval database. (2) The retrieval module uses the database to recommend top K examples (text-label pairs) from source domain data for target domain contents. (3) These examples are adopted as few-shot demonstrations for the inference module to process the target domain content. The RAEmoLLM can effectively enhance the general performance of LLMs in cross-domain misinformation detection tasks through affect-based retrieval, without fine-tuning. We evaluate our framework on three misinformation benchmarks. Results show that RAEmoLLM achieves significant improvements compared to the other few-shot methods on three datasets, with the highest increases of 15.64%, 31.18%, and 15.73% respectively. This project is available at https://github.com/lzw108/RAEmoLLM.