Pub-Guard-LLM is the first LLM-based system specifically designed to detect fraud in biomedical scientific literature, providing reliable explanations for its decisions.

• Offers three deployment modes: vanilla reasoning, retrieval-augmented generation, and multi-agent debate
• Specifically tailored to identify fraudulent practices in biomedical research publications
• Provides transparent explanations alongside detection results, enhancing trust and verification
• Addresses the growing threat of fraudulent publications that undermine medical research credibility

This innovation helps protect the integrity of medical research by identifying problematic publications before they can influence clinical decisions or research directions, ultimately safeguarding patient safety and scientific progress.

Pub-Guard-LLM: Detecting Fraudulent Biomedical Articles with Reliable Explanations