This repository is a curated list of academic papers focusing on the application of Large Language Models (LLMs) and Multimodal LLMs (MLLMs) in misinformation research. It serves as a valuable resource for researchers, academics, and practitioners interested in understanding and leveraging LLMs for detecting, analyzing, and mitigating the spread of misinformation across various domains.

How It Works

The repository categorizes papers based on the role LLMs play in misinformation research: as information providers/data generators/analyzers, as tool users (e.g., accessing search engines), as decision-makers/explainers, and in other related tasks like propagation simulation or correction. This structured approach highlights the diverse capabilities of LLMs in tackling misinformation, from generating synthetic data to providing explanations for detected falsehoods.

Highlighted Details

• Extensive coverage of LLM applications in fake news detection, fact-checking, and rumor analysis.
• Includes papers on multimodal misinformation detection using Vision-Language Models (VLMs).
• Features research on LLM-driven data augmentation, adversarial attacks, and explainability.
• Lists resources like datasets (ARG, MM-Soc) and tutorials on LLM-based misinformation research.

Maintenance & Community

This is a curated list of papers, and the repository itself does not appear to have active development or a dedicated community forum mentioned in the README. It functions as a static reference.

Licensing & Compatibility

The repository itself does not specify a license. The linked papers are subject to their respective publisher's licenses.

Limitations & Caveats

This repository is a bibliography and does not provide any code, models, or tools for misinformation research. Its utility is limited to providing pointers to relevant academic literature.