Can Large Language Models Detect Rumors on Social Media?

Qiang Liu,Xiang Tao,Junfei Wu,Shu Wu,Liang Wang

2024-02-09

Abstract:In this work, we investigate to use Large Language Models (LLMs) for rumor detection on social media. However, it is challenging for LLMs to reason over the entire propagation information on social media, which contains news contents and numerous comments, due to LLMs may not concentrate on key clues in the complex propagation information, and have trouble in reasoning when facing massive and redundant information. Accordingly, we propose an LLM-empowered Rumor Detection (LeRuD) approach, in which we design prompts to teach LLMs to reason over important clues in news and comments, and divide the entire propagation information into a Chain-of-Propagation for reducing LLMs' burden. We conduct extensive experiments on the Twitter and Weibo datasets, and LeRuD outperforms several state-of-the-art rumor detection models by 3.2% to 7.7%. Meanwhile, by applying LLMs, LeRuD requires no data for training, and thus shows more promising rumor detection ability in few-shot or zero-shot scenarios.

Information Retrieval,Computation and Language

What problem does this paper attempt to address?

### What problem does this paper attempt to solve? The main purpose of this paper is to explore how to use large language models (LLMs) to detect rumors on social media. Specifically: 1. **Research Background**: - The development of social media has accelerated the speed of information dissemination, but it has also exacerbated the spread of rumors. - Although large language models (LLMs) have achieved great success, they can also be used to generate a large amount of unidentifiable information, further exacerbating the spread of rumors. 2. **Core Issues**: - How to enable LLMs to effectively reason about complex dissemination information on social media to detect rumors. - Currently, LLMs find it difficult to focus on key clues and struggle to reason effectively when faced with a large number of redundant comments. 3. **Proposed Method**: - The paper proposes a method called LeRuD (LLM-empowered Rumor Detection), which guides LLMs to focus on key clues in news and comments by designing specific prompts. - Dissemination information is broken down into a chronological "propagation chain" so that LLMs can reason step by step. 4. **Experimental Results**: - Extensive experiments were conducted on Twitter and Weibo datasets, and the results show that LeRuD significantly outperforms existing rumor detection models in zero-shot settings. - LeRuD can achieve high-performance rumor detection without training data, making it suitable for few-shot or zero-shot scenarios. In summary, this paper aims to efficiently detect rumors on social media using LLMs and improve the reasoning ability of LLMs by designing reasonable prompts and propagation chain structures.

Can Large Language Models Detect Rumors on Social Media?

Detecting the Rumor Patterns Integrating Features of User, Content, and the Spreading Structure.

Interpretable Graph Neural Network for Social Media Rumor Detection

Rumour Evaluation with Very Large Language Models

Research Status of Deep Learning Methods for Rumor Detection

RP-DNN: A Tweet level propagation context based deep neural networks for early rumor detection in Social Media

Towards Robust Rumor Detection with Graph Contrastive and Curriculum Learning

Zero-Shot Rumor Detection with Propagation Structure via Prompt Learning

Message Injection Attack on Rumor Detection under the Black-Box Evasion Setting Using Large Language Model

Detect Rumors in Microblog Posts for Low-Resource Domains via Adversarial Contrastive Learning

Towards low-resource rumor detection: Unified contrastive transfer with propagation structure

Rumor Detection on Social Media with Event Augmentations

Rumour Detection and Analysis on Twitter

Rumor Mitigation in Social Media Platforms with Deep Reinforcement Learning

CED: Credible Early Detection of Social Media Rumors

Statistical and Semantic Analysis of Rumors in Chinese Social Media

Rumor Detection Based on Representative User Characteristics Learning Through Propagation

Rumor2vec: A rumor detection framework with joint text and propagation structure representation learning

Region-enhanced Deep Graph Convolutional Networks for Rumor Detection

Rumor detection on social media using hierarchically aggregated feature via graph neural networks

Rumor Detection with a novel graph neural network approach